Lambda vs Paperspace
Lambda
Lambda (Lambda Labs)
Lambda wins · 21–9 (16 drawn)
Access connectivity — stories about access connectivity in this arenaAccess connectivity
Stories about access connectivity in this arena
Ide
developerOpen Jupyter or connect my IDE (VS Code/Cursor) to the instance in one step
weight 1 · round to LambdaLambda instances ship with a preinstalled JupyterLab server and SSH access out of the box, letting a developer open Jupyter or SSH in with no extra setup (docs-3, docs-7). However, there's no explicit documentation of a one-step VS Code/Cursor Remote-SSH or dev-container integration — only generic SSH connectivity is mentioned, requiring the developer to manually configure their IDE's remote connection. Missing for 10: explicit VS Code/Cursor connection docs or one-click IDE integration, independent confirmation that IDE remote-attach works smoothly.
- [claimed-docs] “each ODC instance provides a JupyterLab installation for creating and managing Jupyter notebooks”
- [claimed-docs] “You can connect to your On-Demand Cloud (ODC) instances directly through SSH or by using the preinstalled JupyterLab server.”
- [claimed-docs] “On-Demand Cloud (ODC) provides on-demand access to Linux-based, GPU-backed virtual machine instances.”
Paperspacedisputedcontradicted4/10Paperspace ships a native web-based Jupyter Notebooks product and documents SSH access with 'bring your own key, full root access' to Machines, which in principle supports VS Code Remote-SSH style connections, but there is no documentation of a genuine one-step IDE handshake (e.g., no VS Code/Cursor-specific integration, devcontainer, or remote-SSH config generator). A first-hand community report directly contradicts the 'easy SSH' claim, saying getting SSH access to Paperspace Core VMs was 'an uphill battle' with a 'confusing GUI' instead. missing for 10: explicit VS Code/Cursor one-click connect docs, evidence of an SSH config/remote extension workflow, and independent confirmation that SSH setup is actually fast/frictionless.
- [claimed-docs] “Notebooks are a web-based Jupyter IDE with shared persistent storage for long-term development and inter-notebook collaboration, backed by a…”
- [claimed-docs] “You can connect to your Linux or Windows-based machine using the Paperspace console or desktop app using SSH connection.”
- [claimed-docs] “Bring your SSH key and connect directly to your VM with full root access.”
- [claimed-docs] “Root access, connect with SSH Bring your SSH key and connect directly to your VM with full root access.”
- [community] “Their DL virtual servers (Core) are horrible - very slow internet, takes forever to copy datasets in. Getting SSH access is an uphill battle…”
Networking
developerExpose ports to serve applications from my instance and connect instances over private networking
weight 1 · round drawnLambda's docs show port-level control via Firewall rules restricting incoming traffic (docs-11/18) and SSH/JupyterLab access (docs-7), and high-performance private networking (GPUDirect RDMA) exists within 1-Click Clusters and Managed Kubernetes for GPU node interconnect (docs-36, docs-12/33). However, there's no explicit documentation of a general-purpose private networking/VPC feature connecting arbitrary On-Demand instances, nor clear guidance on opening/exposing custom application ports beyond firewall restriction rules. Missing for 10: explicit port-exposure/ingress configuration for serving apps, dedicated private networking (VPC/VLAN) between standard instances outside of 1CC/K8s clusters, and independent confirmation of these networking features working in practice.
- [claimed-docs] “You can restrict incoming traffic to the instances in a workspace... by creating firewall rules on the Firewall page”
- [claimed-docs] “You can restrict incoming traffic to the instances in a workspace...by creating firewall rules on the [Firewall page]”
- [claimed-docs] “You can connect to your On-Demand Cloud (ODC) instances directly through SSH or by using the preinstalled JupyterLab server.”
- [claimed-docs] “1-Click Clusters (1CC) are high-performance clusters composed of both GPU and CPU nodes, featuring 16 to 512 NVIDIA H100 or B200 SXM Tensor …”
- [claimed-docs] “MK8s provides a Kubernetes environment with GPU and InfiniBand (RDMA) support, and shared persistent storage across all nodes in a 1CC.”
- [claimed-docs] “MK8s provides a Kubernetes environment with GPU and InfiniBand (RDMA) support, and shared persistent storage across all nodes in a 1CC. Clus…”
Docs confirm private networking between machines (shared drives across a private network in paperspace-docs-4) and an option to enable a 'private network' or 'public IP address' when creating a machine (paperspace-docs-25), plus SSH/root access to instances (paperspace-docs-12/19/21). However, there is no explicit documentation of exposing arbitrary application ports or firewall/port-forwarding configuration for serving apps beyond SSH; 'Deployments' (containers-as-a-service, paperspace-docs-8) hints at serving models but isn't tied to port exposure or private networking specifics. missing for 10: explicit port-exposure/firewall configuration docs, hands-on confirmation of connecting instances over private network for app traffic, and any independent corroboration of these networking features working as described.
- [claimed-docs] “Shared drives provide storage that is accessible from multiple machines in a private network.”
- [claimed-docs] “When you create a new machine, you choose your machine type, operating system or custom template, disk size, region, authentication, startin…”
- [claimed-docs] “Deployments are containers-as-a-service that let you run container images and serve machine learning models.”
- [claimed-docs] “Bring your SSH key and connect directly to your VM with full root access.”
- [claimed-docs] “Root access, connect with SSH Bring your SSH key and connect directly to your VM with full root access.”
Ssh
developerSSH into my GPU instance with my own keys and get root-level control of the environment
weight 3 · round to LambdaDocs confirm direct SSH access to GPU instances with user-controlled base images (GPU Base minimal Ubuntu image for high control), cloud-init customization, and API/CLI provisioning, implying standard root SSH access typical of cloud VM instances. Community evidence (unofficial CLI/MCP tool) corroborates real-world SSH-based workflows for launching and connecting to instances. Missing for 10: explicit first-party documentation confirming root/sudo privileges once SSH'd in, and independent hands-on confirmation of key management specifics.
- [claimed-docs] “You can connect to your On-Demand Cloud (ODC) instances directly through SSH or by using the preinstalled JupyterLab server.”
- [claimed-docs] “GPU Base: An image based on Ubuntu Server that includes a minimal set of key AI/ML tools and drivers... Use if: You want a minimal working s…”
- [claimed-docs] “you can add launch-time configuration instructions using [cloud-init]”
- [claimed-docs] “the API allows you to use a different base image for your instance”
- [community] “A developer built an unofficial CLI and MCP server for Lambda cloud GPU instances, enabling AI agents to find, launch, and terminate Lambda …”
Paperspacedisputedcontradicted5/10Paperspace's marketing explicitly promises 'bring your own SSH key' with full root access to the VM, and DigitalOcean docs describe SSH-based connection to machines, but a hands-on community report states that SSH access on Paperspace's Core VMs was 'an uphill battle' with a 'pointless virtual console and confusing GUI' instead of easy SSH, contradicting the frictionless claim. missing for 10: independent confirmation that SSH access works smoothly as advertised, and no rebuttal to the community complaint about SSH being hard to obtain.
- [claimed-docs] “Bring your SSH key and connect directly to your VM with full root access.”
- [claimed-docs] “Root access, connect with SSH Bring your SSH key and connect directly to your VM with full root access.”
- [claimed-docs] “Root access, connect with SSH... Bring your SSH key and connect directly to your VM with full root access.”
- [claimed-docs] “You can connect to your Linux or Windows-based machine using the Paperspace console or desktop app using SSH connection.”
- [community] “Their DL virtual servers (Core) are horrible - very slow internet, takes forever to copy datasets in. Getting SSH access is an uphill battle…”
Agenticness — how well agents can access and operate the productAgenticness
How well agents can access and operate the product
Agent access
ai-native userPoint an agent at llms.txt or agent-oriented docs
weight 2 · round to PaperspaceLambdanone0/10A direct probe of https://docs.lambda.ai/llms.txt returned HTTP 404, and there is no other evidence of an llms.txt file or agent-oriented documentation format anywhere in the pack; the OpenAPI spec is a REST API description, not agent-native docs guidance.
- [probe] “PROBE llms.txt: HTTP 404 at https://docs.lambda.ai/llms.txt”
The Paperspace docs live under docs.digitalocean.com, and that domain does serve a working llms.txt (HTTP 200) with a general description of DigitalOcean's documentation corpus, giving agents a discoverable entry point. However, a direct markdown/agent-friendly version of the Paperspace-specific docs page 404s, and there's no evidence the llms.txt specifically indexes or highlights Paperspace content vs. the broader DigitalOcean product catalog. Missing for 10: Paperspace-specific agent-readable docs (e.g. .md endpoint), confirmation llms.txt references Paperspace pages, and any documentation stating this is an intentional agent-onboarding feature.
ai-native userRun the product headlessly / in CI for automation
weight 2 · round to LambdaLambda's On-Demand Cloud exposes a full REST API for programmatically launching/terminating GPU instances, with cloud-init for launch-time automation, confirmed live and key-gated by a runtime probe — enabling headless/CI-driven provisioning of GPU workloads without any UI interaction. Missing for 10: an official first-party CLI/SDK or CI/CD templates (only an unofficial community-built CLI/MCP server exists) and documented CI examples (e.g., GitHub Actions) from Lambda itself.
- [claimed-docs] “You can also launch instances programmatically by using the Lambda Cloud API.”
- [claimed-docs] “You can also launch instances programmatically by using the Lambda Cloud API. For details, see Launch instances”
- [claimed-docs] “If you're launching your instance with the Lambda Cloud API, you can add launch-time configuration instructions using cloud-init”
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create and manage your Lambda Cloud resources.”
- [probe] “PROBE runtime (recorded 2026-09-05): https://cloud.lambda.ai/api/v1/openapi.json serves the full Lambda Cloud API OpenAPI 3.1 spec (title 'L…”
- [community] “A developer built an unofficial CLI and MCP server for Lambda cloud GPU instances, enabling AI agents to find, launch, and terminate Lambda …”
Paperspace exposes a REST API, JS SDK, and a working CLI (pspace) that can create/manage machines, deployments, and templates non-interactively, and Workflows offers pipeline-style automation — enough to script headless usage in CI. However there's no first-party CI-integration guide (e.g. GitHub Actions), no documented service-account/non-interactive auth flow for CI runners, and community feedback flags friction around SSH/API access reliability and product quality, so full CI-native support isn't clearly evidenced. missing for 10: documented CI/CD integration examples, non-interactive/service-account auth for automation, independent confirmation of reliable headless workflows in CI pipelines.
- [claimed-docs] “API keys let you interact with Paperspace through the Core RESTful API, the Core JavaScript SDK, and Gradient command-line utility (CLI).”
- [claimed-docs] “Programmatically manage Paperspace resources using conventional HTTP requests.”
- [claimed-docs] “Install and use the new Paperspace Command Line Interface.”
- [claimed-docs] “Workflows automate machine learning tasks, combining GPU instances with an expressive syntax to generate production-ready machine learning p…”
- [probe] “official CLI documented at https://docs.digitalocean.com/reference/paperspace/pspace/”
- [probe] “PROBE runtime (recorded 2026-09-05): the official installer (`curl -fsSL https://paperspace.com/install.sh | sh`) installed the pspace CLI i…”
- [community] “Their DL virtual servers (Core) are horrible - very slow internet, takes forever to copy datasets in. Getting SSH access is an uphill battle…”
ai-native userConnect an agent via an official MCP server
weight 3 · round drawnLambdanone0/10Only an unofficial, community-built CLI/MCP server for Lambda GPU instances is documented; there is no evidence of an official first-party MCP server from Lambda itself.
- [community] “A developer built an unofficial CLI and MCP server for Lambda cloud GPU instances, enabling AI agents to find, launch, and terminate Lambda …”
ai-native userUse an official CLI
weight 2 · round to PaperspaceLambdanone0/10Evidence shows only a REST API (openapi.json) and cloud-init/SSH access, with no first-party CLI tool documented; the only CLI mentioned is an unofficial third-party CLI/MCP server built by a community developer for AI agents, not an official Lambda offering.
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create and manage your Lambda Cloud resources.”
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create\nand manage your Lambda Cloud resources.”
- [community] “A developer built an unofficial CLI and MCP server for Lambda cloud GPU instances, enabling AI agents to find, launch, and terminate Lambda …”
- [probe] “PROBE runtime (recorded 2026-09-05): https://cloud.lambda.ai/api/v1/openapi.json serves the full Lambda Cloud API OpenAPI 3.1 spec (title 'L…”
Paperspace ships an official CLI (pspace), documented in DigitalOcean docs and independently verified via a runtime probe showing successful install and functional commands for machines, templates, autoscaling, and deployments. missing for 10: independent third-party review specifically praising CLI agentic workflows, and no evidence of AI-native scripting/automation examples beyond basic resource management.
- [claimed-docs] “Install and use the new Paperspace Command Line Interface.”
- [claimed-docs] “API keys let you interact with Paperspace through the Core RESTful API, the Core JavaScript SDK, and Gradient command-line utility (CLI).”
- [probe] “official CLI documented at https://docs.digitalocean.com/reference/paperspace/pspace/”
- [probe] “PROBE runtime (recorded 2026-09-05): the official installer (`curl -fsSL https://paperspace.com/install.sh | sh`) installed the pspace CLI i…”
ai-native userDrive the product through a documented public API
weight 3 · round to LambdaLambda publishes a documented public REST API (Lambda Cloud API) with a live OpenAPI 3.1 spec covering instance provisioning, cluster launch, filesystems, and firewall management, confirmed both in docs and via runtime probe returning the full spec and a key-gated endpoint. Community evidence further shows a third-party built a CLI/MCP server on top of this API enabling AI agents to launch/terminate GPU instances programmatically, corroborating real-world agentic usability. Missing for 10: an official first-party SDK/MCP server and an llms.txt (probe found 404), so slight extra integration work is needed.
- [claimed-docs] “You can also launch instances programmatically by using the Lambda Cloud API.”
- [claimed-docs] “You can also launch instances programmatically by using the Lambda Cloud API. For details, see Launch instances”
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create and manage your Lambda Cloud resources.”
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create\nand manage your Lambda Cloud resources.”
- [probe] “PROBE runtime (recorded 2026-09-05): https://cloud.lambda.ai/api/v1/openapi.json serves the full Lambda Cloud API OpenAPI 3.1 spec (title 'L…”
- [community] “A developer built an unofficial CLI and MCP server for Lambda cloud GPU instances, enabling AI agents to find, launch, and terminate Lambda …”
Paperspace documents a Core RESTful API, JavaScript SDK, and CLI for programmatically managing machines, notebooks, deployments, and workflows, and a runtime probe confirms the official pspace CLI installs and works with commands for machine/deployment management. Missing for 10: no public OpenAPI/swagger spec found, no independent third-party corroboration of API robustness beyond docs.
- [claimed-docs] “API keys let you interact with Paperspace through the Core RESTful API, the Core JavaScript SDK, and Gradient command-line utility (CLI).”
- [claimed-docs] “Programmatically manage Paperspace resources using conventional HTTP requests.”
- [claimed-docs] “Install and use the new Paperspace Command Line Interface.”
- [probe] “official CLI documented at https://docs.digitalocean.com/reference/paperspace/pspace/”
- [probe] “PROBE runtime (recorded 2026-09-05): the official installer (`curl -fsSL https://paperspace.com/install.sh | sh`) installed the pspace CLI i…”
- [probe] “PROBE openapi: all candidate paths 404 (https://docs.digitalocean.com/openapi.json, https://docs.digitalocean.com/swagger.json, https://docs…”
ai-native userIssue scoped/least-privilege API credentials for an agent
weight 2 · round drawnLambdanone0/10Lambda's Cloud API is key-gated (docs-31/39, probe-rt-1 confirms 401 without a key), but there is no evidence of scoped/least-privilege credential issuance (e.g., role-based permissions, read-only vs. write scopes, or per-agent restricted keys) — only that a single API key exists to access all account resources.
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create and manage your Lambda Cloud resources.”
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create\nand manage your Lambda Cloud resources.”
- [probe] “PROBE runtime (recorded 2026-09-05): https://cloud.lambda.ai/api/v1/openapi.json serves the full Lambda Cloud API OpenAPI 3.1 spec (title 'L…”
Paperspacenone0/10Paperspace offers API keys for programmatic access, but evidence shows no scoped or least-privilege permission model — keys appear to be account-level, not fine-grained or agent-specific credentials.
- [claimed-docs] “API keys let you interact with Paperspace through the Core RESTful API, the Core JavaScript SDK, and Gradient command-line utility (CLI).”
- [claimed-docs] “Programmatically manage Paperspace resources using conventional HTTP requests.”
ai-native userBuild against official SDKs
weight 2 · round to PaperspaceLambda documents a public, key-gated REST/Cloud API with a full OpenAPI 3.1 spec that could underlie SDK development, but no official first-party SDK client libraries (Python/JS/etc.) are evidenced — only an unofficial third-party CLI/MCP server exists. Missing for 10: official SDK packages/libraries, first-party language bindings, docs referencing an 'SDK' rather than raw REST endpoints.
- [claimed-docs] “You can also launch instances programmatically by using the Lambda Cloud API.”
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create and manage your Lambda Cloud resources.”
- [probe] “PROBE runtime (recorded 2026-09-05): https://cloud.lambda.ai/api/v1/openapi.json serves the full Lambda Cloud API OpenAPI 3.1 spec (title 'L…”
- [community] “A developer built an unofficial CLI and MCP server for Lambda cloud GPU instances, enabling AI agents to find, launch, and terminate Lambda …”
Docs confirm an official Core RESTful API and Core JavaScript SDK plus a documented CLI (pspace) verified via runtime probe, giving developers concrete official interfaces to build against. However, evidence only names one language SDK (JavaScript), with no mention of Python or other SDKs, and no independent corroboration of SDK quality or ecosystem breadth. missing for 10: evidence of additional language SDKs (e.g. Python), independent developer corroboration of SDK reliability, and a public SDK repo/changelog.
- [claimed-docs] “API keys let you interact with Paperspace through the Core RESTful API, the Core JavaScript SDK, and Gradient command-line utility (CLI).”
- [claimed-docs] “Programmatically manage Paperspace resources using conventional HTTP requests.”
- [claimed-docs] “Install and use the new Paperspace Command Line Interface.”
- [probe] “official CLI documented at https://docs.digitalocean.com/reference/paperspace/pspace/”
- [probe] “PROBE runtime (recorded 2026-09-05): the official installer (`curl -fsSL https://paperspace.com/install.sh | sh`) installed the pspace CLI i…”
ai-native userSubscribe to events via webhooks
weight 2 · round to LambdaLambda's API docs mention webhook notifications only for support ticket events ('Lambda can send webhook notifications to your URL when support ticket events occur'), not for core compute/instance lifecycle events that an AI-native/agentic user would most want to subscribe to. missing for 10: webhook support for instance state changes or job/cluster events, first-party docs on webhook setup/payload schema, and independent confirmation of use in agentic workflows.
- [claimed-docs] “Lambda can send webhook notifications to your URL when support ticket events occur,\nenabling near real-time integration with your systems.”
Paperspacenone0/10No evidence of any webhook or event-subscription mechanism in Paperspace's API, CLI, or docs; only REST API, CLI, and SDK access are documented.
- [claimed-docs] “Programmatically manage Paperspace resources using conventional HTTP requests.”
- [claimed-docs] “Install and use the new Paperspace Command Line Interface.”
- [probe] “official CLI documented at https://docs.digitalocean.com/reference/paperspace/pspace/”
Agentic features
ai-native userSet up automations that run autonomously in the background
weight 2 · round to PaperspaceLambdanone0/10Lambda is GPU IaaS with an API, cloud-init for launch-time config, and webhooks for support tickets, but there is no native scheduler/automation service that runs workflows autonomously in the background; the only agentic automation (CLI/MCP server to launch/terminate instances via AI agents) is an unofficial third-party project, not a Lambda-native feature.
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create and manage your Lambda Cloud resources.”
- [claimed-docs] “Lambda can send webhook notifications to your URL when support ticket events occur,\nenabling near real-time integration with your systems.”
- [claimed-docs] “If you're launching your instance with the Lambda Cloud API, you can add launch-time configuration instructions using cloud-init”
- [community] “A developer built an unofficial CLI and MCP server for Lambda cloud GPU instances, enabling AI agents to find, launch, and terminate Lambda …”
Paperspace's 'Workflows' feature is documented as automating ML tasks/pipelines, and 'Deployments' run containerized models continuously, both of which could constitute background automation, plus API/CLI access enables scripting external automations. However there's no documentation of triggers, schedules, or persistent autonomous execution akin to cron/agentic loops, and no community evidence confirming this works well in practice. Missing for 10: scheduling/trigger mechanisms, evidence of long-running unattended execution, and independent confirmation that Workflows/Deployments actually operate autonomously without manual intervention.
- [claimed-docs] “Workflows automate machine learning tasks, combining GPU instances with an expressive syntax to generate production-ready machine learning p…”
- [claimed-docs] “Deployments are containers-as-a-service that let you run container images and serve machine learning models.”
- [claimed-docs] “API keys let you interact with Paperspace through the Core RESTful API, the Core JavaScript SDK, and Gradient command-line utility (CLI).”
- [claimed-docs] “Install and use the new Paperspace Command Line Interface.”
ai-native userOperate the product with natural-language commands
weight 2 · round to LambdaLambda itself only exposes a REST API (docs-1..41) with no first-party natural-language or agent interface; the only NL-command capability comes from a third-party developer's unofficial CLI/MCP server that lets AI agents launch/terminate instances via commands like 'launch an H100' (lambda-labs-comm-3). This is an extra, unofficial tool rather than a supported product feature. missing for 10: first-party NL/agent interface, official MCP server or chat-based control, documentation of natural-language command support.
- [community] “A developer built an unofficial CLI and MCP server for Lambda cloud GPU instances, enabling AI agents to find, launch, and terminate Lambda …”
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create and manage your Lambda Cloud resources.”
- [probe] “PROBE runtime (recorded 2026-09-05): https://cloud.lambda.ai/api/v1/openapi.json serves the full Lambda Cloud API OpenAPI 3.1 spec (title 'L…”
Api quality
ai-native userExplore an interactive API reference with runnable examples
weight 2 · round drawnLambdanone0/10Lambda exposes a raw OpenAPI 3.1 spec (cloud.lambda.ai/api/v1/openapi.json) and documents REST endpoints, but there is no evidence of an interactive, browsable API reference (e.g., Swagger UI, 'try it out' console) with runnable examples — probes for docs.lambda.ai/openapi.json, swagger.json, and llms.txt all 404.
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create and manage your Lambda Cloud resources.”
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create\nand manage your Lambda Cloud resources.”
- [probe] “PROBE llms.txt: HTTP 404 at https://docs.lambda.ai/llms.txt”
- [probe] “PROBE openapi: all candidate paths 404 (https://docs.lambda.ai/openapi.json, https://docs.lambda.ai/swagger.json, https://docs.lambda.ai/api…”
- [probe] “PROBE runtime (recorded 2026-09-05): https://cloud.lambda.ai/api/v1/openapi.json serves the full Lambda Cloud API OpenAPI 3.1 spec (title 'L…”
Paperspacenone0/10No evidence of an interactive API reference with runnable examples; probes show no OpenAPI/swagger spec found (404s) and no interactive docs playground mentioned, only static docs and CLI/API mentions.
- [probe] “PROBE openapi: all candidate paths 404 (https://docs.digitalocean.com/openapi.json, https://docs.digitalocean.com/swagger.json, https://docs…”
- [probe] “PROBE docs-md: HTTP 404 at https://docs.digitalocean.com/products/paperspace/.md”
- [claimed-docs] “Programmatically manage Paperspace resources using conventional HTTP requests.”
ai-native userDownload a machine-readable API spec (OpenAPI or equivalent)
weight 2 · round to LambdaLambda publishes a live, machine-readable OpenAPI 3.1 spec for its Cloud API at https://cloud.lambda.ai/api/v1/openapi.json, confirmed both by docs references and a runtime probe that fetched the full spec keylessly. Minor gap: the probe found the docs.lambda.ai domain itself doesn't serve openapi.json at the expected conventional path (404s), requiring the correct subdomain. missing for 10: independent third-party confirmation of spec completeness/versioning beyond the runtime probe.
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create and manage your Lambda Cloud resources.”
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create\nand manage your Lambda Cloud resources.”
- [probe] “PROBE runtime (recorded 2026-09-05): https://cloud.lambda.ai/api/v1/openapi.json serves the full Lambda Cloud API OpenAPI 3.1 spec (title 'L…”
- [probe] “PROBE openapi: all candidate paths 404 (https://docs.lambda.ai/openapi.json, https://docs.lambda.ai/swagger.json, https://docs.lambda.ai/api…”
Paperspacenone0/10Paperspace/DigitalOcean documents a Core RESTful API and CLI, but explicit probes for an OpenAPI/Swagger spec at all standard locations returned 404, and no machine-readable spec is referenced anywhere in the docs.
- [claimed-docs] “Programmatically manage Paperspace resources using conventional HTTP requests.”
- [probe] “PROBE openapi: all candidate paths 404 (https://docs.digitalocean.com/openapi.json, https://docs.digitalocean.com/swagger.json, https://docs…”
ai-native userTest against a sandbox environment without touching production data
weight 1 · round drawnLambdanone0/10Lambda's docs describe launching isolated GPU instances, firewalls, and filesystems, but there is no documented sandbox/production separation concept, test-mode, or data-isolation feature aimed at safely testing without touching production data — users would have to build this themselves by manually spinning up separate instances. missing for 10: explicit sandbox/staging environment feature, guidance on isolating test data from production, any mention of a 'sandbox mode' or non-production testing workflow.
- [claimed-docs] “On-Demand Cloud (ODC) provides on-demand access to Linux-based, GPU-backed virtual machine instances.”
- [claimed-docs] “You can restrict incoming traffic to the instances in a workspace... by creating firewall rules on the Firewall page”
- [claimed-docs] “A filesystem is a high-capacity regional file store you can attach to your instance to store datasets and back up system state.”
Paperspacenone0/10Paperspace documentation covers general-purpose VM/notebook/deployment provisioning but never describes a distinct sandbox mode or production-data isolation guarantee; nothing in the evidence ties machine creation to safely testing against non-production data.
- [claimed-docs] “Machines are Linux and Windows virtual machines with persistent storage, GPU options, and free unlimited bandwidth.”
- [claimed-docs] “When you create a new machine, you choose your machine type, operating system or custom template, disk size, region, authentication, startin…”
- [claimed-docs] “Deployments are containers-as-a-service that let you run container images and serve machine learning models.”
ai-native userRely on versioned APIs with a documented deprecation policy
weight 2 · round to LambdaThe Lambda Cloud API is clearly versioned (path-based v1, documented via a live OpenAPI 3.1 spec with rate limits), satisfying the 'versioned APIs' half of the story, but no evidence anywhere in the pack mentions a deprecation policy, versioning changelog, or sunset process for older API versions. Missing for 10: documented deprecation/versioning policy, changelog of breaking changes, migration guidance between API versions, independent confirmation of stability guarantees.
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create and manage your Lambda Cloud resources.”
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create\nand manage your Lambda Cloud resources.”
- [probe] “PROBE runtime (recorded 2026-09-05): https://cloud.lambda.ai/api/v1/openapi.json serves the full Lambda Cloud API OpenAPI 3.1 spec (title 'L…”
Paperspacenone0/10Evidence shows Paperspace has a REST API, CLI, and SDK, but nothing documents API versioning schemes or a deprecation policy anywhere in the docs or community sources.
- [claimed-docs] “API keys let you interact with Paperspace through the Core RESTful API, the Core JavaScript SDK, and Gradient command-line utility (CLI).”
- [claimed-docs] “Programmatically manage Paperspace resources using conventional HTTP requests.”
- [claimed-docs] “Install and use the new Paperspace Command Line Interface.”
Automation depth — how much of the product can run unattendedAutomation depth
How much of the product can run unattended
ai-native userPerform bulk operations across many items at once
weight 2 · round to LambdaLambda supports large-scale provisioning (1-Click Clusters of 16–512+ GPUs, Slurm-based multi-node job scheduling) and a REST API for programmatic instance management, which lets an AI-native user manage many GPU resources at once. However, the documented API is rate-limited to 1 req/s (and launches limited per 12s), with no evidence of true bulk/batch endpoints for creating, updating, or deleting many items in a single call. Missing for 10: explicit bulk-create/bulk-delete API operations, batch job submission tooling beyond Slurm's node scheduling, and evidence of high-throughput automation without rate-limit friction.
- [claimed-docs] “You can also launch instances programmatically by using the Lambda Cloud API.”
- [claimed-docs] “1-Click Clusters (1CC) are high-performance clusters composed of both GPU and CPU nodes, featuring 16 to 512 NVIDIA H100 or B200 SXM Tensor …”
- [claimed-docs] “Production-ready clusters from 16 to 2,000+ NVIDIA B200 or H100 GPUs.”
- [claimed-docs] “Slurm automatically schedules workloads, maximizing cluster utilization while preventing resource contention.”
- [probe] “PROBE runtime (recorded 2026-09-05): https://cloud.lambda.ai/api/v1/openapi.json serves the full Lambda Cloud API OpenAPI 3.1 spec (title 'L…”
Paperspacenone0/10Paperspace exposes an API, CLI (pspace), and Core JS SDK for programmatic resource management, which theoretically could be scripted for bulk actions, but no evidence documents any actual bulk/batch operation feature (e.g., multi-machine create/delete/update in one call, batch endpoints, or CLI loop/apply commands) across many items at once.
- [claimed-docs] “API keys let you interact with Paperspace through the Core RESTful API, the Core JavaScript SDK, and Gradient command-line utility (CLI).”
- [claimed-docs] “Programmatically manage Paperspace resources using conventional HTTP requests.”
- [claimed-docs] “Install and use the new Paperspace Command Line Interface.”
- [probe] “official CLI documented at https://docs.digitalocean.com/reference/paperspace/pspace/”
- [probe] “PROBE runtime (recorded 2026-09-05): the official installer (`curl -fsSL https://paperspace.com/install.sh | sh`) installed the pspace CLI i…”
ai-native userDefine rules that trigger actions automatically on events
weight 3 · round drawnLambdanone0/10Lambda is a GPU IaaS provider; the only event-triggered mechanism found is webhook notifications for support-ticket events (docs-40) and cloud-init instructions applied only at launch time (docs-6, docs-24, docs-32) — neither constitutes a general rules/automation engine letting users define arbitrary triggers-to-actions. No autoscaling policies, alert-based actions, or rule-definition UI/API are documented. missing for 10: a general event-rule engine (define trigger conditions + arbitrary actions), autoscaling/alerting automation, and any first-party support beyond narrow support-ticket webhooks.
- [claimed-docs] “Lambda can send webhook notifications to your URL when support ticket events occur,\nenabling near real-time integration with your systems.”
- [claimed-docs] “you can add launch-time configuration instructions using [cloud-init]”
- [claimed-docs] “If you're launching your instance with the Lambda Cloud API, you can add launch-time configuration instructions using cloud-init”
ai-native userSchedule recurring jobs or workflows
weight 2 · round drawnLambdanone0/10Lambda offers Slurm-based job scheduling for HPC workloads and a REST API with cloud-init for launch-time configuration, but none of the evidence describes a mechanism for scheduling recurring/cron-like jobs or automated recurring workflows — Slurm here is described as scheduling submitted workloads, not recurring automation. missing for 10: any cron/recurring job scheduler, workflow orchestration triggers, or documentation of repeat/interval-based job execution.
- [claimed-docs] “Slurm automatically schedules workloads, maximizing cluster utilization while preventing resource contention.”
- [claimed-docs] “The quick start takes you from SSH to your first multi-node GPU job in minutes.”
- [claimed-docs] “Lambda monitors and maintains the health of Slurm daemons such as slurmctld and slurmdbd.”
- [claimed-docs] “Lambda monitors and maintains the health of Slurm daemons such as `slurmctld` and `slurmdbd`.”
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create and manage your Lambda Cloud resources.”
Paperspacenone0/10Paperspace docs mention on-demand Workflows for ML pipelines and auto-shutdown after inactivity, but there is no evidence of a scheduler, cron-like trigger, or recurring/periodic job execution capability. Absence of evidence for this applicable capability yields none.
- [claimed-docs] “Workflows automate machine learning tasks, combining GPU instances with an expressive syntax to generate production-ready machine learning p…”
- [claimed-docs] “The auto-shutdown feature places your machine in a stop state after a predefined period of inactivity, which can range from one hour to one …”
Capacity availability — stories about capacity availability in this arenaCapacity availability
Stories about capacity availability in this arena
Availability
ml engineerSee real-time GPU availability by type and region before I try to provision, instead of discovering stockouts by failure
weight 2 · round drawnLambdanone0/10The evidence describes Lambda's instance types, API endpoints for launching/managing instances, and pricing, but nowhere shows a real-time GPU availability/capacity view by type and region that an engineer could check before provisioning. Docs even describe access as 'first-come' (docs-28), implying no visibility into stock levels, and no dashboard or API field for capacity-by-region is mentioned.
- [claimed-docs] “ODC offers a variety of predefined instance types to support different workload requirements. Available GPUs include the state-of-the-art NV…”
- [claimed-docs] “Deploy NVIDIA B200, H100, A100, or GH200 instances in minutes with self-serve, first-come access.”
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create and manage your Lambda Cloud resources.”
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create\nand manage your Lambda Cloud resources.”
Paperspacenone0/10No evidence of any real-time GPU availability dashboard, API endpoint, or CLI command showing stock/quota by type and region; docs only describe choosing machine type/region at creation time and needing approval for high-end GPUs, with no visibility into availability before provisioning.
- [claimed-docs] “There are several machine types grouped by CPU, GPU, and multi-GPU. If you choose any high-end machines, such as NVIDIA H100, you need to re…”
- [claimed-docs] “When you create a new machine, you choose your machine type, operating system or custom template, disk size, region, authentication, startin…”
Hardware
ml engineerChoose from current-generation datacenter GPUs (H100/H200/B200 class) as well as cheaper previous-generation options
weight 2 · round to LambdaLambda's docs clearly list current-gen H100/B200 (and GH200) options alongside older A100 in on-demand instances and clusters, covering both current and previous-generation GPU tiers with flexible on-demand access. Missing for 10: no explicit pricing comparison table or independent benchmark confirming actual availability/pricing tiers side-by-side.
- [claimed-docs] “ODC offers a variety of predefined instance types to support different workload requirements. Available GPUs include the state-of-the-art NV…”
- [claimed-docs] “Deploy NVIDIA B200, H100, A100, or GH200 instances in minutes with self-serve, first-come access.”
- [claimed-docs] “1-Click Clusters (1CC) are high-performance clusters composed of both GPU and CPU nodes, featuring 16 to 512 NVIDIA H100 or B200 SXM Tensor …”
- [claimed-docs] “Production-ready clusters from 16 to 2,000+ NVIDIA B200 or H100 GPUs.”
- [community] “Lambda Labs dropped H100 on-demand GPU pricing to $1.99/GPU/Hour (announced via Twitter, June 2023).”
Docs confirm access to current-gen H100 (behind an approval gate) and previous-gen A100 GPUs, plus NVLink support and 'largest GPU catalog' claims, supporting a range of GPU tiers and cost/performance tradeoffs (docs-11, docs-17, docs-20, docs-26). However, there's no explicit mention of H200/B200-class GPUs or a clear listing of older, cheaper GPU tiers (e.g., T4/P100/V100) to fully match the story's specificity. missing for 10: explicit H200/B200 listing, explicit older/cheaper GPU tier catalog, independent benchmarking/corroboration of availability.
- [claimed-docs] “Choose from the largest GPU catalog in the world. Leverage the latest NVIDIA GPUs including Ampere A100s with up to 8 GPUs.”
- [claimed-docs] “There are several machine types grouped by CPU, GPU, and multi-GPU. If you choose any high-end machines, such as NVIDIA H100, you need to re…”
- [claimed-docs] “Enhance data transfer speeds and scalability between NVIDIA GPUs for high-performance computing tasks by enabling NVLink.”
- [claimed-docs] “Easily change instance types anytime so you always have access to the mix of cost and performance. Cancel anytime.”
- [claimed-docs] “Leverage the latest NVIDIA GPUs including Ampere A100s with up to 8 GPUs.”
Quotas
platform engineerSee documented quotas and instance limits and raise them through a defined process
weight 2 · round to PaperspaceLambdanone0/10Evidence shows only generic API rate limits and a vague 'contact us for reserved capacity' pricing note, but no documented per-account instance/GPU quotas nor any defined process (e.g., support ticket workflow, quota dashboard) to request limit increases.
- [claimed-docs] “Contact us for reserved capacity at our lowest prices.”
- [probe] “PROBE runtime (recorded 2026-09-05): https://cloud.lambda.ai/api/v1/openapi.json serves the full Lambda Cloud API OpenAPI 3.1 spec (title 'L…”
Docs mention that high-end GPU machines (e.g., H100) require an approval request, implying some quota/limit gating exists, but there is no dedicated documentation of quota tiers, numeric limits, or a defined escalation/support process for raising them beyond that single mention. Missing for 10: a documented quotas/limits reference page, explicit instance-count or resource caps, and a clear support ticket/process for requesting increases.
- [claimed-docs] “There are several machine types grouped by CPU, GPU, and multi-GPU. If you choose any high-end machines, such as NVIDIA H100, you need to re…”
Clusters scale — stories about clusters scale in this arenaClusters scale
Stories about clusters scale in this arena
Clusters
ml engineerProvision a multi-node GPU cluster with fast interconnect for distributed training without a sales cycle
weight 3 · round to LambdaLambda offers self-serve 1-Click Clusters (16-512 H100/B200 GPUs) with GPUDirect RDMA up to 3200 Gb/s, launched instantly with no long-term commitment and no sales cycle mentioned, plus Managed Slurm/Kubernetes for orchestration and a live self-serve API confirmed by runtime probe. Community evidence corroborates self-serve on-demand access and even third-party tooling built on the API, though no independent hands-on report specifically confirms multi-node cluster provisioning experience. Missing for 10: independent/hands-on validation of the 1-Click Cluster provisioning flow itself and interconnect performance in practice.
- [claimed-docs] “1-Click Clusters (1CC) are high-performance clusters composed of both GPU and CPU nodes, featuring 16 to 512 NVIDIA H100 or B200 SXM Tensor …”
- [claimed-docs] “1-Click Clusters (1CC) are high-performance clusters composed of both GPU and CPU nodes, featuring 16 to 512 NVIDIA H100 or B200 SXM Tensor …”
- [claimed-docs] “The quick start takes you from SSH to your first multi-node GPU job in minutes.”
- [claimed-docs] “Production-ready clusters from 16 to 2,000+ NVIDIA B200 or H100 GPUs.”
- [claimed-docs] “Deploy NVIDIA B200, H100, A100, or GH200 instances in minutes with self-serve, first-come access.”
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create and manage your Lambda Cloud resources.”
- [probe] “PROBE runtime (recorded 2026-09-05): https://cloud.lambda.ai/api/v1/openapi.json serves the full Lambda Cloud API OpenAPI 3.1 spec (title 'L…”
- [community] “A developer built an unofficial CLI and MCP server for Lambda cloud GPU instances, enabling AI agents to find, launch, and terminate Lambda …”
Paperspacenone0/10Evidence shows Paperspace provisions single machines with up to 8 GPUs and NVLink for intra-node scaling, but nothing describes provisioning a multi-node cluster with fast cross-node interconnect (e.g., InfiniBand/RDMA) for distributed training, nor any self-service multi-node cluster workflow.
- [claimed-docs] “Choose from the largest GPU catalog in the world. Leverage the latest NVIDIA GPUs including Ampere A100s with up to 8 GPUs.”
- [claimed-docs] “Enhance data transfer speeds and scalability between NVIDIA GPUs for high-performance computing tasks by enabling NVLink.”
- [claimed-docs] “Leverage the latest NVIDIA GPUs including Ampere A100s with up to 8 GPUs.”
- [claimed-docs] “There are several machine types grouped by CPU, GPU, and multi-GPU. If you choose any high-end machines, such as NVIDIA H100, you need to re…”
Orchestration
platform engineerSchedule jobs on managed Slurm or Kubernetes instead of building my own scheduler on raw nodes
weight 2 · round to LambdaLambda offers Managed Slurm and Managed Kubernetes (MK8s) as first-party managed schedulers on top of 1-Click Clusters, with Slurm handling automatic job scheduling/utilization and daemon health monitoring, and MK8s providing preconfigured GPU/RDMA Kubernetes clusters ready for workload deployment — directly fulfilling the 'managed scheduler instead of building your own' story. Missing for 10: independent/hands-on validation of Slurm or K8s job scheduling in production (community evidence only covers billing and unofficial CLI tooling, not scheduler usage), and no detail on multi-tenancy/queue customization depth.
- [claimed-docs] “MK8s provides a Kubernetes environment with GPU and InfiniBand (RDMA) support, and shared persistent storage across all nodes in a 1CC.”
- [claimed-docs] “Lambda provides support according to the service level agreements (SLAs) in place with the customer.”
- [claimed-docs] “The quick start takes you from SSH to your first multi-node GPU job in minutes.”
- [claimed-docs] “Lambda monitors and maintains the health of Slurm daemons such as slurmctld and slurmdbd.”
- [claimed-docs] “MK8s provides a Kubernetes environment with GPU and InfiniBand (RDMA) support, and shared persistent storage across all nodes in a 1CC. Clus…”
- [claimed-docs] “Lmod for environment management: Use Lmod to dynamically load and unload software modules available on the cluster, such as HPC-X, Node.js, …”
- [claimed-docs] “Slurm automatically schedules workloads, maximizing cluster utilization while preventing resource contention.”
- [claimed-docs] “Lambda monitors and maintains the health of Slurm daemons such as `slurmctld` and `slurmdbd`.”
Openness — open source, data portability, and self-hosting storiesOpenness
Open source, data portability, and self-hosting stories
ai-native userDo everything through the API that I can do in the UI
weight 2 · round drawnLambda publishes a live, documented REST API (OpenAPI spec, key-gated) covering core instance lifecycle actions (launch, list, terminate, cloud-init, custom images) that mirror UI capabilities for On-Demand Cloud instances, and third parties have built agent-facing CLIs/MCP wrappers on top of it. However, evidence doesn't show API parity for other UI-managed features like firewall rules, filesystem management, 1-Click Cluster/Slurm cluster provisioning, or usage/billing dashboards. Missing for 10: documented API endpoints for firewalls, filesystems, 1CC/Slurm cluster lifecycle, and usage/billing views; independent confirmation of full UI-API parity.
- [claimed-docs] “You can also launch instances programmatically by using the Lambda Cloud API.”
- [claimed-docs] “You can also launch instances programmatically by using the Lambda Cloud API. For details, see Launch instances”
- [claimed-docs] “the API allows you to use a different base image for your instance”
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create and manage your Lambda Cloud resources.”
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create\nand manage your Lambda Cloud resources.”
- [probe] “PROBE runtime (recorded 2026-09-05): https://cloud.lambda.ai/api/v1/openapi.json serves the full Lambda Cloud API OpenAPI 3.1 spec (title 'L…”
- [community] “A developer built an unofficial CLI and MCP server for Lambda cloud GPU instances, enabling AI agents to find, launch, and terminate Lambda …”
Paperspace ships both an API and a first-party pspace CLI that support machine creation, custom templates, autoscaling groups, and deployments (paperspace-docs-1, paperspace-docs-6, paperspace-docs-23/24, paperspace-probe-4, paperspace-probe-rt-1), giving strong API/CLI parity for core resource lifecycle tasks. However, some UI-only features like SSH connection setup are described as done via console/desktop app (paperspace-docs-2), no public OpenAPI/Swagger spec was found (paperspace-probe-3), and there's no explicit evidence that shared drives, auto-shutdown, or notebook management are fully API-controllable. Missing for 10: OpenAPI/spec confirmation of full endpoint coverage, explicit API parity for shared drives/auto-shutdown/notebooks, and independent hands-on confirmation that API-driven workflows match UI capability without gaps.
- [claimed-docs] “You can create a Linux machine using the Paperspace console, Paperspace API, or Paperspace CLI.”
- [claimed-docs] “API keys let you interact with Paperspace through the Core RESTful API, the Core JavaScript SDK, and Gradient command-line utility (CLI).”
- [claimed-docs] “Programmatically manage Paperspace resources using conventional HTTP requests.”
- [claimed-docs] “Install and use the new Paperspace Command Line Interface.”
- [probe] “official CLI documented at https://docs.digitalocean.com/reference/paperspace/pspace/”
- [probe] “PROBE runtime (recorded 2026-09-05): the official installer (`curl -fsSL https://paperspace.com/install.sh | sh`) installed the pspace CLI i…”
- [claimed-docs] “You can connect to your Linux or Windows-based machine using the Paperspace console or desktop app using SSH connection.”
- [probe] “PROBE openapi: all candidate paths 404 (https://docs.digitalocean.com/openapi.json, https://docs.digitalocean.com/swagger.json, https://docs…”
ai-native userExport all of my data in open formats and leave
weight 3 · round to LambdaLambda's filesystems support S3-compatible tools like rclone and s5cmd, letting users copy datasets and files out in standard, open formats rather than being locked into a proprietary export mechanism, and the Cloud API allows programmatic access to resource metadata. However, there's no explicit documentation of a full account-data export (billing, usage history, configs) or a one-click 'export everything and leave' workflow. Missing for 10: explicit full-account data export/portability guarantee, documentation of exporting non-file data (billing/usage/API keys), independent verification that egress is unrestricted or free of lock-in fees.
- [claimed-docs] “This adapter allows you to use S3-compatible tools like rclone and s5cmd to copy files to and from your filesystems and perform common file …”
- [claimed-docs] “This adapter allows you to use S3-compatible tools like `rclone` and `s5cmd` to copy files to and from your filesystems”
- [claimed-docs] “A filesystem is a high-capacity regional file store you can attach to your instance to store datasets and back up system state.”
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create and manage your Lambda Cloud resources.”
Paperspace machines use standard SSH/root access and shared drives, and API/CLI access lets users programmatically pull down VM configs, custom templates, and data via SSH/SCP, which provides some data portability. However there is no documented one-click 'export all data' or bulk data-export feature, no explicit open-format guarantee for notebooks/deployments/workflows metadata, and no evidence of a full account data export or migration tool. missing for 10: dedicated bulk data export/download feature, explicit open-format export guarantees for notebooks/deployments/workflows, documented account-level data portability or migration tooling, independent confirmation of successful full data export.
- [claimed-docs] “You can connect to your Linux or Windows-based machine using the Paperspace console or desktop app using SSH connection.”
- [claimed-docs] “Shared drives provide storage that is accessible from multiple machines in a private network.”
- [claimed-docs] “API keys let you interact with Paperspace through the Core RESTful API, the Core JavaScript SDK, and Gradient command-line utility (CLI).”
- [claimed-docs] “Programmatically manage Paperspace resources using conventional HTTP requests.”
- [claimed-docs] “Install and use the new Paperspace Command Line Interface.”
- [probe] “official CLI documented at https://docs.digitalocean.com/reference/paperspace/pspace/”
Pricing billing — stories about pricing billing in this arenaPricing billing
Stories about pricing billing in this arena
Billing
platform engineerPull usage and billing breakdowns programmatically to attribute GPU spend by team or workload
weight 2 · round drawnLambdanone0/10Lambda's docs describe billing by hourly/minute increments and a web 'Usage page' with an Instances tab for viewing monthly usage (lambda-labs-docs-10, lambda-labs-docs-20), but this is UI-only, not a programmatic API. The published Lambda Cloud API (lambda-labs-docs-31/39, confirmed live via probe) covers instance provisioning/management, not billing or usage export, and no team/workload cost-attribution tagging or billing API endpoint is documented anywhere in the evidence.
- [claimed-docs] “ODC prices instances by hourly usage and bills in one-minute increments.”
- [claimed-docs] “To view your monthly usage, navigate to the [Usage page]...and then click the **Instances** tab.”
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create and manage your Lambda Cloud resources.”
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create\nand manage your Lambda Cloud resources.”
- [probe] “PROBE runtime (recorded 2026-09-05): https://cloud.lambda.ai/api/v1/openapi.json serves the full Lambda Cloud API OpenAPI 3.1 spec (title 'L…”
ml engineerI am billed at per-second or per-minute granularity and only while my instance is actually running
weight 3 · round to LambdaLambda's docs explicitly state ODC instances are priced hourly but billed in one-minute increments, and billing only accrues while an instance is running/attached — corroborated by a community report where a user was billed for the full duration their instance remained running (idle counted as running, confirming billing follows instance lifecycle rather than usage activity). Missing for 10: true per-second billing granularity (only per-minute is documented), and independent/user confirmation that billing precisely matches the one-minute increment claim.
- [claimed-docs] “ODC prices instances by hourly usage and bills in one-minute increments.”
- [community] “User left a Lambda Labs GH200 instance running after brief testing and was billed $583 for 391 hours. Lambda's support said: 'Lambda does no…”
Paperspacenone0/10No evidence in the pack states per-second or per-minute billing granularity; the only granularity mentioned is an 'hourly rate' (paperspace-comm-8) and a flat monthly subscription tier with fixed hours (paperspace-comm-5), neither of which confirms sub-minute billing precision while running.
- [community] “I use a Paperspace VM + Parsec for personal ML projects. An hourly rate on a standard VM w/GPU is better than buying a local machine, and yo…”
- [community] “I use Stable Diffusion with Paperspace's Pro tier ($9/mo) for up to 6 hours of non per-usage GPU time, avoiding worry about electricity cost…”
Discovery
ai agentQuery the GPU catalog with live pricing and availability from a public or documented endpoint before committing any spend
weight 2 · round to LambdaLambda publishes a full OpenAPI 3.1 spec for its Cloud API keylessly (lambda-labs-probe-rt-1), and docs confirm the API is used to create/manage instances programmatically (lambda-labs-docs-5, lambda-labs-docs-23, lambda-labs-docs-31/39), with a separate public pricing page (lambda-labs-docs-41). An unofficial MCP/CLI already lets agents launch/terminate Lambda GPUs (lambda-labs-comm-3), implying some catalog query capability exists. However, actual provisioning calls (e.g. /instances) require an API key (401 per probe), and no evidence explicitly shows a documented, unauthenticated 'instance-types/catalog' endpoint returning live pricing+availability distinct from the static marketing pricing page. Missing for 10: explicit documentation of a public catalog/instance-types endpoint with live pricing+availability, and independent confirmation an agent can query it pre-spend without a key.
- [probe] “PROBE runtime (recorded 2026-09-05): https://cloud.lambda.ai/api/v1/openapi.json serves the full Lambda Cloud API OpenAPI 3.1 spec (title 'L…”
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create and manage your Lambda Cloud resources.”
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create\nand manage your Lambda Cloud resources.”
- [claimed-docs] “Clear, straightforward pricing for Instances, 1-Click Clusters™, and Superclusters.”
- [community] “A developer built an unofficial CLI and MCP server for Lambda cloud GPU instances, enabling AI agents to find, launch, and terminate Lambda …”
- [claimed-docs] “You can also launch instances programmatically by using the Lambda Cloud API.”
Paperspacenone0/10While Paperspace documents machine types, GPU options, and an API/CLI for creating machines, there is no evidence of a documented endpoint or CLI command that returns live pricing or availability data for the GPU catalog before provisioning—only marketing claims like 'save up to 70%' and 'largest GPU catalog' with no queryable pricing API surfaced.
- [claimed-docs] “Choose from the largest GPU catalog in the world. Leverage the latest NVIDIA GPUs including Ampere A100s with up to 8 GPUs.”
- [claimed-docs] “There are several machine types grouped by CPU, GPU, and multi-GPU. If you choose any high-end machines, such as NVIDIA H100, you need to re…”
- [claimed-docs] “Programmatically manage Paperspace resources using conventional HTTP requests.”
- [probe] “PROBE openapi: all candidate paths 404 (https://docs.digitalocean.com/openapi.json, https://docs.digitalocean.com/swagger.json, https://docs…”
- [probe] “PROBE runtime (recorded 2026-09-05): the official installer (`curl -fsSL https://paperspace.com/install.sh | sh`) installed the pspace CLI i…”
Pricing
platform engineerLock in reserved or committed-use discounts for sustained GPU capacity
weight 1 · round to LambdaLambda explicitly advertises a path to reserved/committed capacity pricing ('Contact us for reserved capacity at our lowest prices') and also offers Private Cloud for large single-tenant multi-year deployments, confirming the capability exists. However, this is sales-assisted only — there's no self-service reservation mechanism, no published discount tiers/terms, and on-demand pages emphasize 'No long-term commitments,' contrasting with the reserved offering. Missing for 10: self-service reservation/commitment workflow, published discount rates or contract terms, and independent evidence of actual reserved pricing outcomes.
- [claimed-docs] “Contact us for reserved capacity at our lowest prices.”
- [claimed-docs] “Private Cloud is the ideal solution for organizations requiring a single-tenant cluster with 1,000+ GPUs. Private Cloud customers have low-l…”
- [claimed-docs] “Launch 16x to 512x NVIDIA H100 or B200 GPU clusters instantly, or start with a single-node instance. No long-term commitments.”
Paperspacenone0/10No evidence of reserved instances, committed-use contracts, or sustained-use discounts; only pay-as-you-go pricing claims like 'save up to 70%' and 'cancel anytime' are documented, which implies on-demand rather than committed pricing.
- [claimed-docs] “Save up to 70% on compute costs”
- [claimed-docs] “Easily change instance types anytime so you always have access to the mix of cost and performance. Cancel anytime.”
ml engineerSee the published per-GPU-hour price for every GPU type on a public pricing page without talking to sales
weight 3 · round to LambdaLambda publishes a public pricing page (lambda.ai/pricing) with clear per-GPU-hour pricing and explicitly states 'Clear, straightforward pricing for Instances, 1-Click Clusters, and Superclusters' without requiring sales contact for on-demand instances; community evidence corroborates historical public per-GPU pricing announcements (e.g., $1.99/GPU/Hour H100). Missing for 10: independent third-party confirmation that ALL current GPU types' per-hour rates are listed publicly (reserved/private cloud capacity explicitly requires contacting sales per docs-35), and no direct evidence snippet showing the actual price table contents.
- [claimed-docs] “Clear, straightforward pricing for Instances, 1-Click Clusters™, and Superclusters.”
- [claimed-docs] “Production-ready clusters from 16 to 2,000+ NVIDIA B200 or H100 GPUs.”
- [claimed-docs] “Contact us for reserved capacity at our lowest prices.”
- [community] “Lambda Labs dropped H100 on-demand GPU pricing to $1.99/GPU/Hour (announced via Twitter, June 2023).”
Paperspacenone0/10The evidence pack contains marketing claims like 'save up to 70% on compute costs' but no actual published per-GPU-hour pricing table or pricing page content is shown; no citation demonstrates a public price list for each GPU type.
- [claimed-docs] “Save up to 70% on compute costs”
- [claimed-docs] “Save up to 70% on compute costs Spend significantly less on your GPU compute compared to the major public clouds or buying your own servers.”
- [claimed-docs] “Save up to 70% on compute costs. Spend significantly less on your GPU compute compared to the major public clouds or buying your own servers…”
Spot
ml engineerRent spot or interruptible GPU capacity at a deep discount with clearly documented preemption semantics
weight 3 · round drawnLambdanone0/10Lambda's docs describe only on-demand, reserved, and private-cloud pricing tiers; there is no mention of spot/interruptible/preemptible instances or any preemption semantics. The billing evidence even highlights that Lambda bills continuously for any running instance regardless of use, reinforcing the absence of a spot/interruptible model.
- [claimed-docs] “ODC prices instances by hourly usage and bills in one-minute increments.”
- [claimed-docs] “Contact us for reserved capacity at our lowest prices.”
- [claimed-docs] “Clear, straightforward pricing for Instances, 1-Click Clusters™, and Superclusters.”
- [community] “User left a Lambda Labs GH200 instance running after brief testing and was billed $583 for 391 hours. Lambda's support said: 'Lambda does no…”
Privacy posture — data-handling and privacy storiesPrivacy posture
Data-handling and privacy stories
ai-native userChoose where my data is stored (region/residency)
weight 2 · round to PaperspaceLambdanone0/10Lambda's docs mention that filesystems are a 'regional' store (implying infrastructure regions exist) but provide no evidence of user-facing controls to select a specific region or data-residency zone for compliance/privacy purposes. Missing for 10: explicit region-selection UI/API, documented list of available regions, and any data-residency/compliance guarantees.
- [claimed-docs] “A filesystem is a high-capacity regional file store you can attach to your instance to store datasets and back up system state.”
- [claimed-docs] “A _filesystem_ is a high-capacity regional file store you can attach to your instance to store datasets and back up system state.”
Docs confirm you select a 'region' when creating a machine ([paperspace-docs-25]), showing basic region choice for compute placement, but there is no evidence of dedicated data residency controls, storage region selection, compliance certifications (e.g., GDPR, SOC2 data residency), or documentation on where notebooks/deployments data is stored. missing for 10: explicit data residency/compliance documentation, storage-specific region selection, list of available regions, independent confirmation of enforcement.
- [claimed-docs] “When you create a new machine, you choose your machine type, operating system or custom template, disk size, region, authentication, startin…”
ai-native userControl data retention and deletion
weight 2 · round drawnLambdanone0/10Lambda is a GPU cloud infrastructure provider (compute instances, clusters, filesystems, API); there is no evidence of any data retention/deletion controls, data lifecycle policies, or privacy dashboard for AI-native users to manage what data is stored or deleted. The evidence covers filesystems for storage and billing/usage tracking but nothing about retention limits or deletion mechanisms. missing for 10: documented data retention policy, deletion/opt-out controls, data lifecycle settings, any privacy-posture documentation.
- [claimed-docs] “A filesystem is a high-capacity regional file store you can attach to your instance to store datasets and back up system state.”
- [claimed-docs] “A _filesystem_ is a high-capacity regional file store you can attach to your instance to store datasets and back up system state.”
- [claimed-docs] “To view your monthly usage, navigate to the [Usage page]...and then click the **Instances** tab.”
Paperspacenone0/10No evidence pack items address data retention policies, deletion controls, or privacy/data lifecycle management for Paperspace resources like notebooks, storage, or machine data; the docs cover creation, connection, and compute features but nothing on retention/deletion control.
Provisioning lifecycle — creating, updating, and tearing down resources across their lifecycleProvisioning lifecycle
Creating, updating, and tearing down resources across their lifecycle
Agent ops
ai agentProvision a GPU, monitor it, run a workload, and tear it down end to end through documented APIs, CLI, or MCP without a human in the console
weight 3 · round to LambdaLambda documents a full REST API for provisioning, launching, and managing on-demand instances/clusters (with cloud-init for workload setup and a live, key-gated OpenAPI endpoint confirmed by probe), plus usage/monitoring pages and GPU dashboards — enabling non-human, API-driven lifecycle management. However, there is no official CLI and no first-party MCP server; the only MCP integration found is a third-party/unofficial community-built CLI+MCP wrapper, and monitoring is oriented toward billing/usage dashboards rather than a documented job-status API for agents. missing for 10: official CLI, first-party MCP server, and a documented workload-monitoring/job-status API distinct from usage billing.
- [claimed-docs] “You can also launch instances programmatically by using the Lambda Cloud API.”
- [claimed-docs] “You can also launch instances programmatically by using the Lambda Cloud API. For details, see Launch instances”
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create and manage your Lambda Cloud resources.”
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create\nand manage your Lambda Cloud resources.”
- [probe] “PROBE runtime (recorded 2026-09-05): https://cloud.lambda.ai/api/v1/openapi.json serves the full Lambda Cloud API OpenAPI 3.1 spec (title 'L…”
- [claimed-docs] “you can add launch-time configuration instructions using [cloud-init]”
- [claimed-docs] “To view your monthly usage, navigate to the [Usage page]...and then click the **Instances** tab.”
- [community] “A developer built an unofficial CLI and MCP server for Lambda cloud GPU instances, enabling AI agents to find, launch, and terminate Lambda …”
Paperspace documents API keys, a Core RESTful API, and an official CLI (pspace) for creating/managing machines, custom templates, and auto-shutdown, and a runtime probe confirms the CLI installs and lists machine/deployment commands — enough to plausibly script provisioning and teardown without a human in the console. However there is no MCP server, no documented end-to-end 'create→monitor→run workload→teardown' workflow example, and community reports describe friction getting SSH/API access working reliably on Core VMs, undercutting a clean automated experience. Missing for 10: an MCP server, explicit workload-run/monitoring API docs, and independent confirmation of a full agent-driven lifecycle without console fallback.
- [claimed-docs] “You can create a Linux machine using the Paperspace console, Paperspace API, or Paperspace CLI.”
- [claimed-docs] “API keys let you interact with Paperspace through the Core RESTful API, the Core JavaScript SDK, and Gradient command-line utility (CLI).”
- [claimed-docs] “Programmatically manage Paperspace resources using conventional HTTP requests.”
- [claimed-docs] “Install and use the new Paperspace Command Line Interface.”
- [claimed-docs] “The auto-shutdown feature places your machine in a stop state after a predefined period of inactivity, which can range from one hour to one …”
- [probe] “official CLI documented at https://docs.digitalocean.com/reference/paperspace/pspace/”
- [probe] “PROBE runtime (recorded 2026-09-05): the official installer (`curl -fsSL https://paperspace.com/install.sh | sh`) installed the pspace CLI i…”
- [community] “Their DL virtual servers (Core) are horrible - very slow internet, takes forever to copy datasets in. Getting SSH access is an uphill battle…”
Manage
ml engineerSet auto-shutdown timers or spend limits so a forgotten instance can't silently run up a huge bill
weight 2 · round to PaperspaceLambdanone0/10No documentation or feature evidence shows auto-shutdown timers, idle detection, or spend-limit controls; on the contrary, community evidence explicitly states Lambda 'does not distinguish between idle and in use instance states' and a user was billed $583 for 391 hours of an idle GH200 instance, confirming the absence of this safeguard.
- [community] “User left a Lambda Labs GH200 instance running after brief testing and was billed $583 for 391 hours. Lambda's support said: 'Lambda does no…”
Paperspace documents an auto-shutdown feature that stops a machine after inactivity (from 1 hour to 1 week), directly addressing the 'forgotten instance' risk, and this is corroborated by official docs. However, there is no evidence of configurable spend/budget limits or billing caps, which is the other half of the story. missing for 10: documented spend-limit/budget-cap feature, independent hands-on confirmation that auto-shutdown reliably prevents runaway billing.
- [claimed-docs] “The auto-shutdown feature places your machine in a stop state after a predefined period of inactivity, which can range from one hour to one …”
developerStart, stop, restart, and terminate instances programmatically and keep paying only for what is running
weight 2 · round to PaperspaceLambdadisputedcontradicted6/10Lambda's API and docs support programmatic launch/terminate of instances with per-minute billing (lambda-labs-docs-5, -10, -23, -31, -39, runtime probe confirms live REST API), but there's no documented stop/start (pause) capability distinct from terminate, and a hands-on community report shows billing continues for idle-but-running instances regardless of usage, contradicting 'pay only for what is running' in the sense of active workload — user was billed $583 for an idle instance because Lambda 'does not distinguish between idle and in use instance states' (lambda-labs-comm-1). missing for 10: documented stop/pause (vs. terminate) lifecycle action, confirmation that idle time is excluded from billing, independent corroboration beyond one HN anecdote.
- [claimed-docs] “You can also launch instances programmatically by using the Lambda Cloud API.”
- [claimed-docs] “ODC prices instances by hourly usage and bills in one-minute increments.”
- [claimed-docs] “You can also launch instances programmatically by using the Lambda Cloud API. For details, see Launch instances”
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create and manage your Lambda Cloud resources.”
- [probe] “PROBE runtime (recorded 2026-09-05): https://cloud.lambda.ai/api/v1/openapi.json serves the full Lambda Cloud API OpenAPI 3.1 spec (title 'L…”
- [community] “User left a Lambda Labs GH200 instance running after brief testing and was billed $583 for 391 hours. Lambda's support said: 'Lambda does no…”
Evidence confirms programmatic creation of machines via API/CLI/console and an auto-shutdown feature that stops machines after inactivity to control costs, and the pspace CLI clearly manages 'machine' resources. However, no direct documentation shows explicit start/stop/restart/terminate commands or billing-only-when-running semantics beyond auto-shutdown. missing for 10: explicit API/CLI start, stop, restart, terminate operations, confirmation that billing pauses immediately on stop, independent verification of lifecycle commands working.
- [claimed-docs] “You can create a Linux machine using the Paperspace console, Paperspace API, or Paperspace CLI.”
- [claimed-docs] “The auto-shutdown feature places your machine in a stop state after a predefined period of inactivity, which can range from one hour to one …”
- [claimed-docs] “API keys let you interact with Paperspace through the Core RESTful API, the Core JavaScript SDK, and Gradient command-line utility (CLI).”
- [claimed-docs] “Programmatically manage Paperspace resources using conventional HTTP requests.”
- [claimed-docs] “Install and use the new Paperspace Command Line Interface.”
- [probe] “official CLI documented at https://docs.digitalocean.com/reference/paperspace/pspace/”
- [probe] “PROBE runtime (recorded 2026-09-05): the official installer (`curl -fsSL https://paperspace.com/install.sh | sh`) installed the pspace CLI i…”
Provision
developerProvision an on-demand GPU instance from the console or API and be running code on it within minutes
weight 3 · round to LambdaLambda's docs and live API confirm on-demand GPU provisioning via console or REST API, with cloud-init, SSH, and JupyterLab access enabling immediate code execution, and pricing pages claim deployment 'in minutes.' A community-built CLI/MCP tool also independently confirms real users launching and SSHing into instances quickly. Missing for 10: independent benchmark/timing evidence validating the 'minutes' claim and first-hand confirmation of no friction in provisioning flow.
- [claimed-docs] “Launch 16x to 512x NVIDIA H100 or B200 GPU clusters instantly, or start with a single-node instance. No long-term commitments.”
- [claimed-docs] “You can also launch instances programmatically by using the Lambda Cloud API.”
- [claimed-docs] “you can add launch-time configuration instructions using [cloud-init]”
- [claimed-docs] “You can connect to your On-Demand Cloud (ODC) instances directly through SSH or by using the preinstalled JupyterLab server.”
- [claimed-docs] “On-Demand Cloud (ODC) provides on-demand access to Linux-based, GPU-backed virtual machine instances.”
- [claimed-docs] “Deploy NVIDIA B200, H100, A100, or GH200 instances in minutes with self-serve, first-come access.”
- [probe] “PROBE runtime (recorded 2026-09-05): https://cloud.lambda.ai/api/v1/openapi.json serves the full Lambda Cloud API OpenAPI 3.1 spec (title 'L…”
- [community] “A developer built an unofficial CLI and MCP server for Lambda cloud GPU instances, enabling AI agents to find, launch, and terminate Lambda …”
Paperspacedisputedcontradicted5/10Docs and marketing show provisioning via console/API/CLI, SSH root access, and ML-ready templates enabling fast startup (paperspace-docs-1,6,10,12,13,24), and a runtime probe confirms the CLI installs and lists machine-management commands (paperspace-probe-rt-1). However, a hands-on community report directly contradicts the 'connect and run in minutes' claim: 'Getting SSH access is an uphill battle; instead there's a pointless virtual console... very slow internet, takes forever to copy datasets in' (paperspace-comm-7), and another user calls the notebook experience 'really bad' (paperspace-comm-6). Missing for 10: independent verified timing benchmarks of provisioning-to-running-code, resolution of the SSH-access friction reported by users, and more recent hands-on corroboration.
- [claimed-docs] “You can create a Linux machine using the Paperspace console, Paperspace API, or Paperspace CLI.”
- [claimed-docs] “API keys let you interact with Paperspace through the Core RESTful API, the Core JavaScript SDK, and Gradient command-line utility (CLI).”
- [claimed-docs] “Go from signup to training a model in seconds”
- [claimed-docs] “Bring your SSH key and connect directly to your VM with full root access.”
- [claimed-docs] “Choose "ML in a Box" template that comes preinstalled with all the major ML frameworks and CUDA® drivers.”
- [claimed-docs] “Install and use the new Paperspace Command Line Interface.”
- [probe] “PROBE runtime (recorded 2026-09-05): the official installer (`curl -fsSL https://paperspace.com/install.sh | sh`) installed the pspace CLI i…”
- [community] “Their DL virtual servers (Core) are horrible - very slow internet, takes forever to copy datasets in. Getting SSH access is an uphill battle…”
- [community] “PaperSpace product recently has been really bad. Gradient Notebooks are a worse version of Google Colab, useless for serious DL product buil…”
Serverless endpoints — stories about serverless endpoints in this arenaServerless endpoints
Stories about serverless endpoints in this arena
Serverless
developerDeploy code to autoscaling serverless GPU workers that scale to zero, instead of managing always-on instances
weight 2 · round drawnLambdanone0/10Lambda's documented products are on-demand VMs, 1-Click Clusters, Managed Kubernetes, and Managed Slurm — all provisioned/billed as running instances, not autoscaling serverless GPU functions. Community evidence explicitly confirms Lambda 'does not distinguish between idle and in use instance states' and bills continuously even when idle, the opposite of scale-to-zero behavior. No docs mention a serverless endpoint product or autoscale-to-zero deployment model.
- [claimed-docs] “On-Demand Cloud (ODC) provides on-demand access to Linux-based, GPU-backed virtual machine instances.”
- [claimed-docs] “ODC prices instances by hourly usage and bills in one-minute increments.”
- [community] “User left a Lambda Labs GH200 instance running after brief testing and was billed $583 for 391 hours. Lambda's support said: 'Lambda does no…”
Paperspacenone0/10Paperspace's evidence describes VM-based 'Machines' with auto-shutdown after inactivity (docs-5) and container 'Deployments'/autoscaling-groups (docs-8, probe-rt-1), but nothing confirms true serverless GPU workers that scale to zero per-request and back up automatically without idle billing — the model described is always-provisioned instances that stop, not ephemeral serverless invocation.
- [claimed-docs] “The auto-shutdown feature places your machine in a stop state after a predefined period of inactivity, which can range from one hour to one …”
- [claimed-docs] “Deployments are containers-as-a-service that let you run container images and serve machine learning models.”
- [probe] “PROBE runtime (recorded 2026-09-05): the official installer (`curl -fsSL https://paperspace.com/install.sh | sh`) installed the pspace CLI i…”
Storage data — storing and moving data — persistence, formats, durabilityStorage data
Storing and moving data — persistence, formats, durability
Data movement
developerMove data in and out efficiently — S3-compatible endpoints, cloud-storage sync, or documented transfer tooling
weight 2 · round to LambdaLambda documents an S3-compatible adapter for its filesystems, explicitly supporting rclone and s5cmd for copying data in/out, plus a REST Cloud API and cloud-init for programmatic/automated data and instance management. Missing for 10: no independent hands-on verification of transfer performance/reliability and no native cloud-storage sync service beyond the S3-compatible adapter.
- [claimed-docs] “This adapter allows you to use S3-compatible tools like rclone and s5cmd to copy files to and from your filesystems and perform common file …”
- [claimed-docs] “This adapter allows you to use S3-compatible tools like `rclone` and `s5cmd` to copy files to and from your filesystems”
- [claimed-docs] “A filesystem is a high-capacity regional file store you can attach to your instance to store datasets and back up system state.”
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create and manage your Lambda Cloud resources.”
- [probe] “PROBE runtime (recorded 2026-09-05): https://cloud.lambda.ai/api/v1/openapi.json serves the full Lambda Cloud API OpenAPI 3.1 spec (title 'L…”
Paperspacenone0/10No evidence of S3-compatible endpoints, cloud-storage sync, or dedicated data-transfer tooling; only generic 'shared drives' (docs-4) and CLI/API for resource management are documented, and community reports (paperspace-comm-7) describe data transfer to Core VMs as slow and cumbersome, further undermining any efficiency claim. missing for 10: S3-compatible endpoint, documented sync/transfer tool, benchmarked transfer speeds, and any first-party guidance on moving data in/out.
- [claimed-docs] “Shared drives provide storage that is accessible from multiple machines in a private network.”
- [claimed-docs] “Machines are Linux and Windows virtual machines with persistent storage, GPU options, and free unlimited bandwidth.”
- [community] “Their DL virtual servers (Core) are horrible - very slow internet, takes forever to copy datasets in. Getting SSH access is an uphill battle…”
Storage
developerAttach persistent network storage that survives instance teardown, so datasets and checkpoints outlive any single GPU rental
weight 3 · round to LambdaLambda's docs explicitly describe attachable persistent filesystems ('high-capacity regional file store you can attach to your instance to store datasets and back up system state') that exist independently of any instance, plus S3-compatible tooling (rclone/s5cmd) for moving data in/out, directly matching the story of datasets/checkpoints surviving GPU teardown. missing for 10: no independent/hands-on confirmation of persistence across teardown or details on durability guarantees/pricing for the filesystem.
- [claimed-docs] “A filesystem is a high-capacity regional file store you can attach to your instance to store datasets and back up system state.”
- [claimed-docs] “A _filesystem_ is a high-capacity regional file store you can attach to your instance to store datasets and back up system state.”
- [claimed-docs] “This adapter allows you to use S3-compatible tools like rclone and s5cmd to copy files to and from your filesystems and perform common file …”
- [claimed-docs] “This adapter allows you to use S3-compatible tools like `rclone` and `s5cmd` to copy files to and from your filesystems”
Docs describe persistent storage attached to machines (paperspace-docs-16) and separate 'shared drives' accessible from multiple machines in a private network (paperspace-docs-4), plus persistent notebook storage (paperspace-docs-7), which together imply data can outlive a single GPU instance. However, no evidence explicitly confirms shared drives survive full instance deletion/teardown, nor documents attach/detach workflow, pricing, or size limits, and one community comment complains about slow data transfer into Paperspace VMs (paperspace-comm-7). Missing for 10: explicit lifecycle documentation proving storage persists after instance termination, attach/detach mechanics, and independent hands-on confirmation of durability across teardown.
- [claimed-docs] “Shared drives provide storage that is accessible from multiple machines in a private network.”
- [claimed-docs] “Machines are Linux and Windows virtual machines with persistent storage, GPU options, and free unlimited bandwidth.”
- [claimed-docs] “Notebooks are a web-based Jupyter IDE with shared persistent storage for long-term development and inter-notebook collaboration, backed by a…”
- [community] “Their DL virtual servers (Core) are horrible - very slow internet, takes forever to copy datasets in. Getting SSH access is an uphill battle…”
Templates images — stories about templates images in this arenaTemplates images
Stories about templates images in this arena
Images
developerRun my own Docker image or custom machine template with my exact environment
weight 2 · round to PaperspaceLambda's docs show real support for custom environments: the Cloud API lets you launch instances with 'a different base image' (custom machine template) and cloud-init for launch-time configuration, plus GPU Base images for full control over the Python environment. Containerized workloads (i.e., running an actual Docker image) are supported via Managed Kubernetes (MK8s) rather than as a first-class ODC instance feature. Missing for 10: explicit first-party documentation of directly launching a Docker image as an ODC instance (vs. VM base image), a custom-image/AMI gallery, and independent hands-on confirmation that custom base images work as claimed.
- [claimed-docs] “the API allows you to use a different base image for your instance”
- [claimed-docs] “you can add launch-time configuration instructions using [cloud-init]”
- [claimed-docs] “If you're launching your instance with the Lambda Cloud API, you can add launch-time configuration instructions using cloud-init”
- [claimed-docs] “GPU Base: An image based on Ubuntu Server that includes a minimal set of key AI/ML tools and drivers... Use if: You want a minimal working s…”
- [claimed-docs] “MK8s provides a Kubernetes environment with GPU and InfiniBand (RDMA) support, and shared persistent storage across all nodes in a 1CC.”
- [claimed-docs] “MK8s provides a Kubernetes environment with GPU and InfiniBand (RDMA) support, and shared persistent storage across all nodes in a 1CC. Clus…”
Paperspace lets users create custom machine templates (paperspace-docs-3, paperspace-docs-25) and run container images via Deployments (paperspace-docs-8), plus SSH root access to VMs for custom setups (paperspace-docs-12, paperspace-docs-19). Community reports (paperspace-comm-7) note friction getting reliable SSH/console access on Core VMs, which tempers confidence. Missing for 10: independent hands-on confirmation that custom Docker images run smoothly end-to-end, and clearer detail on template versioning/sharing.
- [claimed-docs] “Custom templates are templates of your machines and their configurations. You can use these custom templates to create new machines with you…”
- [claimed-docs] “Deployments are containers-as-a-service that let you run container images and serve machine learning models.”
- [claimed-docs] “When you create a new machine, you choose your machine type, operating system or custom template, disk size, region, authentication, startin…”
- [claimed-docs] “Bring your SSH key and connect directly to your VM with full root access.”
- [claimed-docs] “Root access, connect with SSH Bring your SSH key and connect directly to your VM with full root access.”
- [community] “Their DL virtual servers (Core) are horrible - very slow internet, takes forever to copy datasets in. Getting SSH access is an uphill battle…”
Templates
developerLaunch from pre-built ML templates (PyTorch, CUDA, vLLM, ComfyUI) instead of assembling an environment from scratch
weight 2 · round drawnLambda's ODC instances preinstall 'Lambda Stack' (AI/ML drivers, tools, frameworks) and offer a 'GPU Base' minimal image, which covers PyTorch/CUDA-style ready environments, plus JupyterLab for quick start. However, there's no evidence of named pre-built templates for vLLM or ComfyUI specifically, only generic 'AI/ML tools and frameworks'. missing for 10: explicit vLLM template/image, explicit ComfyUI template/image, a documented template gallery/catalog beyond Lambda Stack and GPU Base.
- [claimed-docs] “Lambda also preinstalls Lambda Stack, a standard set of AI/ML-related drivers, tools, and frameworks, on the instance”
- [claimed-docs] “GPU Base: An image based on Ubuntu Server that includes a minimal set of key AI/ML tools and drivers.”
- [claimed-docs] “GPU Base: An image based on Ubuntu Server that includes a minimal set of key AI/ML tools and drivers... Use if: You want a minimal working s…”
- [claimed-docs] “each ODC instance provides a JupyterLab installation for creating and managing Jupyter notebooks”
Paperspace docs confirm a pre-built "ML in a Box" template with major ML frameworks and CUDA drivers preinstalled, plus custom-template creation for reuse, supporting the general concept of launching from ready-made ML environments. However, no evidence names specific frameworks like PyTorch, vLLM, or ComfyUI templates, and community comments describe friction (slow setup, SDK upgrade walls, GUI confusion) that undercuts a seamless 'launch instantly' experience. missing for 10: explicit PyTorch/vLLM/ComfyUI-named templates, hands-on confirmation of frictionless template launch, independent corroboration beyond marketing docs.
- [claimed-docs] “Choose "ML in a Box" template that comes preinstalled with all the major ML frameworks and CUDA® drivers.”
- [claimed-docs] “Custom templates are templates of your machines and their configurations. You can use these custom templates to create new machines with you…”
- [claimed-docs] “When you create a new machine, you choose your machine type, operating system or custom template, disk size, region, authentication, startin…”
- [community] “Paperspace gives free access to Graphcore IPU nodes (4 IPUs each), theoretically more throughput than Colab T4. But porting Stable Diffusion…”
- [community] “Their DL virtual servers (Core) are horrible - very slow internet, takes forever to copy datasets in. Getting SSH access is an uphill battle…”
Trust governance — stories about trust governance in this arenaTrust governance
Stories about trust governance in this arena
Compliance
platform engineerVerify the provider's security and compliance posture (SOC 2, data handling, datacenter tiers) before putting proprietary models on it
weight 2 · round drawnLambdanone0/10No evidence in the pack addresses SOC 2 certification, compliance attestations, data handling policies, or datacenter tier ratings for Lambda's cloud offerings; the docs focus entirely on GPU provisioning, clusters, filesystems, and API usage. This is a fair, applicable question for any cloud infrastructure provider hosting proprietary models, but no supporting documentation exists in the evidence pack.
Paperspacenone0/10No evidence pack items mention SOC 2 compliance, certifications, data handling policies, or datacenter tier information; the evidence is entirely product feature docs and community complaints about performance/support, none of which address compliance posture. missing for 10: SOC 2/ISO certifications, data handling/privacy documentation, datacenter tier specs, any compliance attestations.
Governance
platform engineerManage team members with roles and scoped API keys so credentials and spend stay controlled
weight 1 · round drawnLambdanone0/10The evidence pack documents instance provisioning, filesystems, firewalls, and a key-gated REST API, but contains no mention of team/member management, role-based access control, or scoped/restricted API keys for spend control. The billing complaint (lambda-labs-comm-1) actually highlights a lack of granular usage controls rather than confirming governance features.
- [claimed-docs] “The Lambda Cloud API provides a set of REST API endpoints you can use to create and manage your Lambda Cloud resources.”
- [probe] “PROBE runtime (recorded 2026-09-05): https://cloud.lambda.ai/api/v1/openapi.json serves the full Lambda Cloud API OpenAPI 3.1 spec (title 'L…”
- [community] “User left a Lambda Labs GH200 instance running after brief testing and was billed $583 for 391 hours. Lambda's support said: 'Lambda does no…”
Paperspacenone0/10Evidence confirms API keys exist for programmatic access (paperspace-docs-6, paperspace-docs-23) but there is no mention of team member management, role assignment, scoped/permissioned API keys, or spend-control governance features anywhere in the pack.
- [claimed-docs] “API keys let you interact with Paperspace through the Core RESTful API, the Core JavaScript SDK, and Gradient command-line utility (CLI).”
- [claimed-docs] “Programmatically manage Paperspace resources using conventional HTTP requests.”
Not comparable on these axes
ai-native userPlug MCP servers into this product so it can use their tools
weight 3 · not comparableLambdan/aLambda is a GPU cloud/infrastructure platform (instance provisioning, clusters, storage, Slurm/K8s) with no agent or assistant component that could consume external tools via MCP. The evidence even shows the reverse: a third-party built an MCP server *around* Lambda's API so other agents could control Lambda, not Lambda acting as an MCP client itself, so this axis is a category error for this product type.
- [community] “A developer built an unofficial CLI and MCP server for Lambda cloud GPU instances, enabling AI agents to find, launch, and terminate Lambda …”
ai-native userGet AI-generated insights and suggestions from my data inside the product
weight 2 · not comparableLambdan/aLambda is a GPU cloud infrastructure provider (compute instances, clusters, Slurm, Kubernetes) — it is not a data/analytics product that generates AI-driven insights or suggestions from a user's data. This story asks about an application-layer AI-insights feature, which is a category mismatch for an infrastructure/IaaS product; the only AI-related community item is an unofficial third-party MCP/CLI wrapper for provisioning instances, not data insights.
Paperspacen/aPaperspace is an infrastructure/GPU-cloud platform for provisioning machines, notebooks, and deployments; it does not itself analyze user data to generate AI-driven insights or suggestions inside the product. This capability is a category error for an infrastructure/compute provider rather than an analytics or AI-assistant product.
ai-native userDelegate tasks to a built-in AI assistant inside the product
weight 3 · not comparableLambdan/aLambda is a GPU cloud/infrastructure provider (on-demand instances, clusters, Slurm, Kubernetes), not a product with a built-in AI assistant for task delegation. This axis is a category error for an IaaS GPU platform; the community mention of an unofficial MCP/CLI wrapper built by a third party does not constitute a first-party built-in assistant.
ai-native userVersion, review, and roll back my automations
weight 1 · not comparableLambdan/aLambda is a GPU cloud infrastructure/compute provider, not an automation/workflow-builder product; versioning, reviewing, and rolling back 'automations' is a category mismatch — this is a wrong-axis question for an IaaS/GPU platform.
Paperspacen/aPaperspace is a GPU cloud/VM/notebook infrastructure product, not an automation/agent-workflow builder; there is no concept of 'automations' with version history, review, or rollback in its evidence (Machines, Notebooks, Deployments, Workflows are infra features, not user-authored automations with versioning/review workflows). This is a category mismatch rather than a missing feature.
ai-native userRead the product's source under an open license
weight 2 · not comparableLambdan/aLambda is a GPU cloud/infrastructure product, not an open-source software project; there is no indication its core product source code is published under an open license. This axis (reading product source under an open license) is a category mismatch for a cloud compute service rather than something the evidence contradicts.
ai-native userSelf-host the core product
weight 3 · not comparableLambdan/aLambda is a cloud GPU IaaS/compute platform, not open-source software; self-hosting the 'core product' is a category error since the product itself is the hosted cloud infrastructure being rented out, not a deployable application.
ai-native userPrevent my data from being used to train AI models
weight 3 · not comparableLambdan/aLambda is a GPU cloud infrastructure provider (IaaS), not an AI model/chat product that trains models on user data; data-training opt-out is a category error for this kind of product.
ai-native userOpt out of telemetry and usage tracking
weight 2 · not comparableLambdan/aLambda is a GPU cloud/infrastructure provider, not an AI assistant or telemetry-collecting client tool; opting out of telemetry/usage tracking is not a relevant axis for this product category — it's about billing usage of cloud resources, not AI-native telemetry.