Together AI vs Morph
Together AI wins · 22–4 (19 drawn)
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 drawnTogether AI publishes a working llms.txt (HTTP 200) with a documentation index, per-page .md docs, and even a live MCP server for docs lookup, all independently confirmed via probes plus first-party docs on agent-skills. missing for 10: no independent third-party report of an agent actually consuming llms.txt successfully in practice.
- [probe] “PROBE llms.txt: HTTP 200 at https://docs.together.ai/llms.txt # Together AI docs > Documentation for the Together AI platform for inference…”
- [probe] “PROBE docs-md: HTTP 200 at https://docs.together.ai/intro.md > ## Documentation Index > Fetch the complete documentation index at: https://d…”
- [probe] “PROBE mcp-endpoint (2026-09-04): POST initialize to https://docs.together.ai/mcp answered HTTP 200 with a JSON-RPC/MCP response (event: mess…”
- [claimed-docs] “Make your AI coding agent Together-AI-aware with ready-made skills for code generation and an MCP server for live docs lookup.”
- [claimed-docs] “Docs MCP server: Gives your agent live access to this documentation site so it can look up current information without leaving your editor.”
Morph confirms llms.txt is live and documented (HTTP 200) with a full documentation index, and also exposes .md versions of docs pages explicitly for agent consumption, directly matching the story's ask. Missing for 10: no independent/community confirmation of agents actually consuming llms.txt in practice.
- [probe] “PROBE llms.txt: HTTP 200 at https://docs.morphllm.com/llms.txt # Morph API Documentation > https://docs.morphllm.com > Base URL: https://api…”
- [probe] “PROBE docs-md: HTTP 200 at https://docs.morphllm.com/introduction.md > ## Documentation Index > Fetch the complete documentation index at: h…”
- [claimed-docs] “Point any OpenAI SDK at `https://api.morphllm.com/v1`. One API key covers the open-weight chat models and the specialized tools”
ai-native userRun the product headlessly / in CI for automation
weight 2 · round to Together AITogether AI is an API/SDK-first inference platform with OpenAI-compatible REST endpoints, official server-side SDKs (TypeScript/Python), batch inference, streaming via SSE, and structured/function-calling outputs — all natively usable in headless CI/automation pipelines without any UI dependency. Live probes confirm the API and status endpoints are reachable programmatically. missing for 10: no explicit CI/CD example or GitHub Actions template in evidence, no documented CLI exit-code/scripting conventions, and no independent case study of a CI pipeline using Together AI headlessly.
- [claimed-docs] “you can point it at models hosted on Together with two changes: the API key and base URL”
- [claimed-docs] “Call 100+ open-source models with per-token pricing and no provisioning latency.”
- [claimed-docs] “Run asynchronous batch workloads at up to 50% lower cost.”
- [github] “We provide support for streaming responses using Server Sent Events (SSE).”
- [github] “This library provides convenient access to the Together REST API from server-side TypeScript or JavaScript.”
- [probe] “PROBE models-endpoint (2026-09-04): GET https://api.together.xyz/v1/models without a key returned HTTP 401 (Missing API key) — the OpenAI-st…”
- [claimed-docs] “Call chat, image, audio, embedding, and more through one API.”
Morph is delivered as an OpenAI/Anthropic-compatible HTTP API (morph-docs-1, morph-docs-16) plus a batch mode explicitly designed for running 'thousands of chat completions offline' (morph-docs-13), both of which are naturally scriptable/headless and usable from CI pipelines. However, there is no explicit CI/CD documentation, GitHub Actions example, or automation-pipeline guide in the evidence pack. Missing for 10: explicit CI/automation guide or example, dedicated headless-mode docs beyond generic API access, independent confirmation of CI usage.
- [claimed-docs] “Point any OpenAI SDK at `https://api.morphllm.com/v1`. One API key covers the open-weight chat models and the specialized tools”
- [claimed-docs] “Run thousands of chat completions offline at half price.”
- [claimed-docs] “Anthropic Messages supports the same open source chat models, token billing, and rate limits.”
- [probe] “PROBE llms.txt: HTTP 200 at https://docs.morphllm.com/llms.txt # Morph API Documentation > https://docs.morphllm.com > Base URL: https://api…”
ai-native userConnect an agent via an official MCP server
weight 3 · round to Together AITogether AI documents an official Docs MCP server that gives agents live access to documentation, and a live probe confirms the MCP endpoint at docs.together.ai/mcp responds correctly to an initialize call with proper protocol capabilities. This is corroborated by both first-party docs and independent hands-on verification. Missing for 10: broader coverage beyond docs lookup (e.g., an MCP server exposing inference/model-management actions rather than just documentation retrieval), and no community discussion confirming real-world agent integration experiences.
- [claimed-docs] “Make your AI coding agent Together-AI-aware with ready-made skills for code generation and an MCP server for live docs lookup.”
- [claimed-docs] “12 domain-specific skills that load on demand and teach your agent how to write correct Together AI code (right model IDs, SDK patterns, bes…”
- [claimed-docs] “Gives your agent live access to this documentation site so it can look up current information without leaving your editor.”
- [claimed-docs] “Docs MCP server: Gives your agent live access to this documentation site so it can look up current information without leaving your editor.”
- [probe] “official MCP server documented at https://docs.together.ai/docs/agent-skills”
- [probe] “PROBE mcp-endpoint (2026-09-04): POST initialize to https://docs.together.ai/mcp answered HTTP 200 with a JSON-RPC/MCP response (event: mess…”
Morph publishes an official MCP server (guides/mcp.md, quickstart 'Install the Morph MCP so Claude Code can read it') that lets external agents like Claude Code connect to its fast file-editing/search tools, confirmed live via probe [morph-probe-4]. missing for 10: independent/hands-on confirmation that the MCP server works as documented (community evidence only covers the apply model, not MCP connectivity), and no detail on MCP tool list/config beyond the guide.
- [claimed-docs] “Connect your favorite AI tools to Morph's blazing-fast file editing via Model Context Protocol.”
- [claimed-docs] “Install the Morph MCP so Claude Code can read it”
- [claimed-docs] “Use Morph's Warp-Grep for fast, local code search alongside your MCP setup.”
- [probe] “official MCP server documented at https://docs.morphllm.com/guides/mcp”
ai-native userUse an official CLI
weight 2 · round to Together AIDocs mention an official CLI (the 'tg' command, e.g. 'tg beta endpoints deploy...') and explicitly state fine-tuning jobs can be launched 'through the API/SDK, or with the CLI', confirming a real CLI exists for agentic/dev workflows. However, evidence lacks a dedicated CLI reference page, install instructions, full command coverage, or independent hands-on corroboration. Missing for 10: dedicated CLI documentation/installation guide, comprehensive command reference, independent user verification of CLI usage.
- [claimed-docs] “tg beta endpoints deploy google/gemma-4-E4B-it --endpoint my-endpoint”
- [claimed-docs] “You can launch a fine-tuning job from the console, through the API/SDK, or with the CLI”
Morphnone0/10Morph is API/SDK-focused (OpenAI-compatible endpoints, MCP integration, Fast Apply, etc.) but no evidence pack item mentions an official Morph CLI tool; community only references third-party CLIs (llm.datasette.io) connecting to Morph's API, not a first-party CLI shipped by Morph.
ai-native userDrive the product through a documented public API
weight 3 · round to Together AITogether AI exposes an OpenAI-compatible REST API with official SDKs (TypeScript/JS), documented endpoints for chat, batch, fine-tuning, dedicated deployment, and a live, publicly reachable models endpoint confirmed by probe; docs are extensive and public. Missing for 10: a discoverable formal OpenAPI/swagger spec (probe found all candidate paths 404).
- [claimed-docs] “you can point it at models hosted on Together with two changes: the API key and base URL”
- [claimed-docs] “Call 100+ open-source models with per-token pricing and no provisioning latency.”
- [github] “This library provides convenient access to the Together REST API from server-side TypeScript or JavaScript.”
- [probe] “PROBE models-endpoint (2026-09-04): GET https://api.together.xyz/v1/models without a key returned HTTP 401 (Missing API key) — the OpenAI-st…”
- [probe] “PROBE openapi: all candidate paths 404 (https://docs.together.ai/openapi.json, https://docs.together.ai/swagger.json, https://docs.together.…”
Morph documents a public OpenAI-compatible API at api.morphllm.com/v1 plus a native Anthropic Messages endpoint, with SDK docs covering chat, Fast Apply, Compact, Batch, caching, and dedicated endpoints, and this is corroborated by third-party usage (HN commenters, github demos) actually driving it programmatically. Missing for 10: no discoverable OpenAPI/swagger spec (404s on all probed paths) and mixed community reports on output reliability reduce polish.
- [claimed-docs] “Point any OpenAI SDK at `https://api.morphllm.com/v1`. One API key covers the open-weight chat models and the specialized tools”
- [claimed-docs] “Fast Apply merges just the changed lines at 10,500 tok/s, 98% accuracy.”
- [claimed-docs] “Drop filler from chat history and code context at 33,000 tok/s. 50-70% reduction, every surviving line byte-for-byte identical to input.”
- [claimed-docs] “Run thousands of chat completions offline at half price.”
- [claimed-docs] “Anthropic Messages supports the same open source chat models, token billing, and rate limits.”
- [probe] “PROBE llms.txt: HTTP 200 at https://docs.morphllm.com/llms.txt # Morph API Documentation > https://docs.morphllm.com > Base URL: https://api…”
- [probe] “PROBE openapi: all candidate paths 404 (https://docs.morphllm.com/openapi.json, https://docs.morphllm.com/swagger.json, https://docs.morphll…”
- [github] “Use Morph's API to intelligently modify existing code at 4,500+ tokens/sec”
- [community] “Have been using morph for a while (I am one of the authors of goose) and was surprised when introduced at the boost it gave me (much less it…”
- [community] “This uses an OpenAI-compatible endpoint, so got this working with my llm.datasette.io CLI tool... One thing that worries me: since it's usin…”
ai-native userIssue scoped/least-privilege API credentials for an agent
weight 2 · round drawnTogether AInone0/10No evidence of scoped or least-privilege API key issuance (e.g., per-project keys, role/permission scoping, or restricted-scope tokens for agents); docs only mention a single API key used for authentication, with no mention of scoping controls.
- [claimed-docs] “you can point it at models hosted on Together with two changes: the API key and base URL”
Morphnone0/10Evidence shows only a single API key covering all models/tools with no mention of scoped, least-privilege, or role-based credential issuance for agents; no docs on permission scoping, sub-keys, or credential restriction. missing for 10: scoped/limited-permission API key creation, per-agent credential issuance, role-based access control docs.
- [claimed-docs] “Point any OpenAI SDK at `https://api.morphllm.com/v1`. One API key covers the open-weight chat models and the specialized tools”
ai-native userBuild against official SDKs
weight 2 · round to Together AITogether AI provides an official TypeScript/JavaScript SDK (github repo) and OpenAI-compatible API/SDK usage, with docs covering chat, structured outputs, function calling, streaming, fine-tuning, and dedicated deployment all launchable via API/SDK. This directly supports AI-native, agentic development against official SDKs, including agent loop patterns and tool calling. Missing for 10: independent hands-on developer reviews specifically praising SDK quality/completeness, and no official Python SDK repo cited in evidence pack.
- [github] “We provide support for streaming responses using Server Sent Events (SSE).”
- [github] “This library provides convenient access to the Together REST API from server-side TypeScript or JavaScript.”
- [claimed-docs] “you can point it at models hosted on Together with two changes: the API key and base URL”
- [claimed-docs] “Function calling (also called tool calling) lets LLMs respond with structured function names and arguments that you can execute in your appl…”
- [claimed-docs] “To build agent loops, chain tool calls inside one response (multi-step), and conversations that thread tools across many turns (multi-turn).”
- [claimed-docs] “You can launch a fine-tuning job from the console, through the API/SDK, or with the CLI”
- [claimed-docs] “Supported models can return JSON that conforms to any schema you supply, so you can read the output directly in code without retries or frag…”
Morph documents OpenAI-SDK and Anthropic Messages-API compatibility (point any OpenAI SDK at api.morphllm.com, native /v1/messages support), and community evidence confirms real developers wiring it up via OpenAI-compatible clients (e.g. llm.datasette.io CLI). However there's no evidence of a first-party Morph-branded SDK package, and an OpenAPI spec probe returned 404s across all candidate paths, suggesting no formal machine-readable API spec is published. missing for 10: dedicated first-party Morph SDK package, discoverable OpenAPI/swagger spec, broader language-SDK coverage beyond OpenAI/Anthropic compatibility.
- [claimed-docs] “Point any OpenAI SDK at `https://api.morphllm.com/v1`. One API key covers the open-weight chat models and the specialized tools”
- [claimed-docs] “Claude Code speaks the Anthropic Messages API, which Morph serves natively at `/v1/messages`.”
- [claimed-docs] “Anthropic Messages supports the same open source chat models, token billing, and rate limits.”
- [community] “This uses an OpenAI-compatible endpoint, so got this working with my llm.datasette.io CLI tool... One thing that worries me: since it's usin…”
- [probe] “PROBE openapi: all candidate paths 404 (https://docs.morphllm.com/openapi.json, https://docs.morphllm.com/swagger.json, https://docs.morphll…”
ai-native userSubscribe to events via webhooks
weight 2 · round drawnTogether AInone0/10No evidence of webhook subscription support anywhere in the docs, SDKs, or probes; Together AI offers SSE streaming and a docs MCP server, but nothing about webhook event subscriptions for async notifications (e.g., fine-tune job completion, batch job status).
Agentic features
ai-native userSet up automations that run autonomously in the background
weight 2 · round drawnTogether AInone0/10The axis applies to this product kind (peer products hold positive or none verdicts on this story), so lack of evidence for an applicable capability is "none", never "na". (na/none harmonized at arena bring-up — see pipeline/scripts/na-harmonize.ts.)
Morphnone0/10Morph is presented as an LLM inference/API platform (fast apply, compact, batch completions, canary/reflex model switching) rather than a background automation or agent-orchestration product; nothing in the evidence describes setting up autonomous, self-running background automations or scheduled agentic tasks. The closest features (batch processing, canary traffic routing) are developer-triggered infra operations, not autonomous background automations.
- [claimed-docs] “Run thousands of chat completions offline at half price.”
- [claimed-docs] “Move a production agent off Anthropic or OpenAI onto Kimi K3 — a 5% traffic trial or a full migration, gated by Reflexes.”
- [claimed-docs] “A canary keeps your current provider, sends a slice of production traffic to Kimi K3, and compares Reflex results before you switch.”
Api quality
ai-native userExplore an interactive API reference with runnable examples
weight 2 · round drawnTogether AInone0/10The evidence pack shows extensive prose documentation (guides, overview pages) but no reference to an interactive API reference or runnable/try-it-now code examples; a direct probe for OpenAPI/Swagger specs at standard paths returned 404 for all candidates, indicating no interactive spec-driven reference was found. No community or hands-on evidence confirms a runnable API console exists.
- [probe] “PROBE openapi: all candidate paths 404 (https://docs.together.ai/openapi.json, https://docs.together.ai/swagger.json, https://docs.together.…”
- [probe] “PROBE llms.txt: HTTP 200 at https://docs.together.ai/llms.txt # Together AI docs > Documentation for the Together AI platform for inference…”
- [probe] “PROBE docs-md: HTTP 200 at https://docs.together.ai/intro.md > ## Documentation Index > Fetch the complete documentation index at: https://d…”
Morphnone0/10The evidence shows standard markdown docs (llms.txt, introduction.md) but no interactive API reference or runnable-example explorer — probes for an OpenAPI/Swagger spec returned 404 on every candidate path, and no docs mention a live API console. A community mention of a separate 'playground' relates to testing the apply model, not an API reference tool, and doesn't establish this capability.
- [probe] “PROBE openapi: all candidate paths 404 (https://docs.morphllm.com/openapi.json, https://docs.morphllm.com/swagger.json, https://docs.morphll…”
- [probe] “PROBE llms.txt: HTTP 200 at https://docs.morphllm.com/llms.txt # Morph API Documentation > https://docs.morphllm.com > Base URL: https://api…”
- [probe] “PROBE docs-md: HTTP 200 at https://docs.morphllm.com/introduction.md > ## Documentation Index > Fetch the complete documentation index at: h…”
ai-native userDownload a machine-readable API spec (OpenAPI or equivalent)
weight 2 · round drawnTogether AInone0/10Active probing for standard OpenAPI/swagger spec locations (openapi.json, swagger.json, etc.) all returned 404, and no docs page in the pack references a downloadable OpenAPI/machine-readable spec for the Together API.
- [probe] “PROBE openapi: all candidate paths 404 (https://docs.together.ai/openapi.json, https://docs.together.ai/swagger.json, https://docs.together.…”
Morphnone0/10Morph is API-based (OpenAI-compatible endpoint), so an OpenAPI spec is a fair expectation, but a direct probe found all candidate OpenAPI/swagger spec paths returned 404, and no evidence pack item shows a downloadable machine-readable spec.
- [probe] “PROBE openapi: all candidate paths 404 (https://docs.morphllm.com/openapi.json, https://docs.morphllm.com/swagger.json, https://docs.morphll…”
ai-native userRely on versioned APIs with a documented deprecation policy
weight 2 · round drawnTogether AInone0/10The evidence shows a documented, OpenAI-compatible REST API (with a /v1/ endpoint) but contains no mention of a versioning scheme or a documented deprecation policy for API changes. Absence of evidence for this applicable capability means it cannot be credited as delivered.
Morphnone0/10No evidence of API versioning scheme (e.g., v1/v2) or a documented deprecation policy anywhere in the docs; the /v1 path is just the base URL, not a versioning/deprecation commitment. OpenAPI spec probes also 404, further indicating no formal versioned API contract is published.
- [probe] “PROBE openapi: all candidate paths 404 (https://docs.morphllm.com/openapi.json, https://docs.morphllm.com/swagger.json, https://docs.morphll…”
- [claimed-docs] “Point any OpenAI SDK at `https://api.morphllm.com/v1`. One API key covers the open-weight chat models and the specialized tools”
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 Together AITogether AI offers an asynchronous Batch API explicitly designed for bulk workloads across many items at once, at up to 50% lower cost than synchronous calls, alongside dedicated endpoints and fine-tuning pipelines that can process large datasets end-to-end. missing for 10: no documented batch size/throughput limits, no independent/hands-on report validating batch job behavior at scale, and no detail on input format flexibility for large item counts.
- [claimed-docs] “Run asynchronous batch workloads at up to 50% lower cost.”
- [claimed-docs] “Together AI handles the full lifecycle: data upload, training, hosting, and inference on a dedicated endpoint.”
- [claimed-docs] “Fine-tuning tailors a pretrained model to a smaller, targeted dataset so it performs better on a specific task or domain. Together AI handle…”
Morph offers batch offline completions (morph-docs-13) and fast bulk file-editing/apply capabilities (morph-docs-11, morph-gh-1), which support some bulk/automation-depth use cases, but there is no documented feature for bulk operations across many arbitrary items (e.g., batch file renames, mass repo-wide refactors, multi-item dataset operations) beyond chat completions and code apply. missing for 10: explicit bulk/batch operation APIs beyond chat completions, evidence of bulk operations across many discrete items (files, records, tickets) rather than just token throughput, independent verification of bulk reliability at scale.
- [claimed-docs] “Run thousands of chat completions offline at half price.”
- [claimed-docs] “Fast Apply merges just the changed lines at 10,500 tok/s, 98% accuracy.”
- [github] “Use Morph's API to intelligently modify existing code at 4,500+ tokens/sec”
ai-native userSchedule recurring jobs or workflows
weight 2 · round drawnTogether AInone0/10Together AI documents async batch workloads, fine-tuning jobs, and dedicated endpoints, but nowhere in the evidence is there a scheduler, cron-like trigger, or recurring-job/workflow orchestration mechanism; jobs appear to be triggered manually via API/CLI/console rather than on a recurring schedule.
- [claimed-docs] “Run asynchronous batch workloads at up to 50% lower cost.”
- [claimed-docs] “You can launch a fine-tuning job from the console, through the API/SDK, or with the CLI”
- [claimed-docs] “Fine-tuning tailors a pretrained model to a smaller, targeted dataset so it performs better on a specific task or domain. Together AI handle…”
Batch async — stories about batch async in this arenaBatch async
Stories about batch async in this arena
Batch
ml-engineerSubmit asynchronous batch inference jobs at a documented discount versus real-time pricing
weight 2 · round to Together AITogether AI has documented batch inference at up to 50% lower cost than real-time serverless pricing, directly matching the story's requirement for asynchronous jobs at a documented discount [together-ai-docs-3]. missing for 10: independent/hands-on corroboration of actual batch job submission and discount realization beyond the docs page, and more detail on submission workflow/API specifics.
- [claimed-docs] “Run asynchronous batch workloads at up to 50% lower cost.”
Morph documents a batch component explicitly for offline, asynchronous chat completions at 'half price' (morph-docs-13), directly matching the story's async-job-with-discount claim. However, there's no detail on job submission/status API mechanics, SLA/turnaround time, or independent confirmation of the discount in practice. missing for 10: detailed batch API docs (submission, polling, output retrieval), documented turnaround SLAs, independent/community verification of discount pricing.
- [claimed-docs] “Run thousands of chat completions offline at half price.”
Dedicated capacity — stories about dedicated capacity in this arenaDedicated capacity
Stories about dedicated capacity in this arena
Dedicated
ml-engineerDeploy a model on dedicated GPU capacity with autoscaling so my traffic is isolated from the shared serverless pool
weight 2 · round to Together AIDocs explicitly describe Dedicated Model Inference (DMI) for serving models on reserved/isolated hardware distinct from the shared serverless pool, with no hard rate limits, and a dedicated scaling doc showing configurable autoscaling via replica limits, plus deployment of fine-tuned or custom models via CLI. This directly matches the story's requirements of dedicated GPU capacity, autoscaling, and isolation from shared serverless traffic. Missing for 10: independent/hands-on verification of autoscaling behavior in production and concrete GPU-type/SLA details beyond docs.
- [claimed-docs] “Dedicated model inference (DMI) lets you serve a model on reserved hardware, providing several advantages over serverless models”
- [claimed-docs] “Configure your deployment to scale automatically by setting limits on how many replicas it can run.”
- [claimed-docs] “Fine-tuned models: Deploy a model you fine-tuned from a supported base model.”
- [claimed-docs] “tg beta endpoints deploy google/gemma-4-E4B-it --endpoint my-endpoint”
- [claimed-docs] “Dedicated model inference (DMI) lets you serve a model on reserved hardware... Better performance... No hard rate limits”
Morph docs mention 'dedicated endpoints' where you reserve model capacity and plan and Morph provisions/operates it, which implies isolated capacity, but there is no mention of autoscaling, traffic isolation guarantees, or GPU-level specifics. missing for 10: explicit autoscaling behavior, confirmation of traffic isolation from shared serverless pool, GPU capacity details, independent corroboration of dedicated endpoint performance.
- [claimed-docs] “Reserve model capacity by choosing a model and plan. Morph provisions and operates it”
Fine tune serving — stories about fine tune serving in this arenaFine tune serving
Stories about fine tune serving in this arena
Fine tune
ml-engineerFine-tune a supported base model on my own data and serve the result on the same platform
weight 2 · round to Together AIDocs clearly describe fine-tuning (LoRA or full) on supported base models via console/API/CLI, with Together AI handling the full lifecycle including data upload, training, hosting, and inference on a dedicated endpoint, and dedicated endpoints explicitly support deploying fine-tuned models. This is a first-party documented end-to-end workflow with concrete CLI examples. Missing for 10: independent/hands-on confirmation from a third party that a fine-tune-then-serve workflow works as described.
- [claimed-docs] “Fine-tune models on your own data and deploy them for inference.”
- [claimed-docs] “LoRA: Trains a small set of adapter weights on top of the frozen base model... Full fine-tuning: Updates every weight in the base model.”
- [claimed-docs] “You can launch a fine-tuning job from the console, through the API/SDK, or with the CLI”
- [claimed-docs] “Fine-tuned models: Deploy a model you fine-tuned from a supported base model.”
- [claimed-docs] “Together AI handles the full lifecycle: data upload, training, hosting, and inference on a dedicated endpoint.”
- [claimed-docs] “Fine-tuning tailors a pretrained model to a smaller, targeted dataset so it performs better on a specific task or domain. Together AI handle…”
Morph offers 'Reflexes' — small custom classifiers that train in ~30s on your data and are then served/gated in production (morph-docs-6, morph-docs-3, morph-docs-4) — plus dedicated endpoints to reserve capacity for a chosen model (morph-docs-15). This is a narrow, specialized form of 'train on your data and serve it' rather than fine-tuning a general-purpose base chat/completion model. Missing for 10: explicit fine-tuning API/pipeline for full base language models (not just small classifiers), documentation on custom dataset upload/training for chat models, and evidence that a fine-tuned base model (vs. a Reflex classifier) can be served alongside the standard endpoints.
- [claimed-docs] “A small Reflex trains in about 30 seconds.”
- [claimed-docs] “Move a production agent off Anthropic or OpenAI onto Kimi K3 — a 5% traffic trial or a full migration, gated by Reflexes.”
- [claimed-docs] “A canary keeps your current provider, sends a slice of production traffic to Kimi K3, and compares Reflex results before you switch.”
- [claimed-docs] “Reserve model capacity by choosing a model and plan. Morph provisions and operates it”
ml-engineerUpload and serve my own custom model weights or LoRA adapters
weight 2 · round to Together AITogether AI's docs clearly support fine-tuning (LoRA or full) and deploying the resulting fine-tuned models to dedicated inference endpoints, covering the LoRA-adapter serving use case end-to-end (data upload, training, hosting, inference). However, there's no evidence of uploading externally-trained/custom weights not produced via Together's own fine-tuning pipeline (e.g., bringing your own checkpoint from another framework). missing for 10: evidence of uploading externally-trained custom weights/checkpoints (not produced by Together's fine-tuning), and independent/hands-on confirmation of this workflow.
- [claimed-docs] “LoRA: Trains a small set of adapter weights on top of the frozen base model... Full fine-tuning: Updates every weight in the base model.”
- [claimed-docs] “Fine-tuned models: Deploy a model you fine-tuned from a supported base model.”
- [claimed-docs] “Together AI handles the full lifecycle: data upload, training, hosting, and inference on a dedicated endpoint.”
- [claimed-docs] “Fine-tuning tailors a pretrained model to a smaller, targeted dataset so it performs better on a specific task or domain. Together AI handle…”
- [claimed-docs] “You can launch a fine-tuning job from the console, through the API/SDK, or with the CLI”
Morphnone0/10Morph's docs describe serving open-weight chat models, dedicated capacity reservations, and training small in-house 'Reflex' classifiers, but nothing about ml-engineers uploading their own custom model weights or LoRA adapters for serving. missing for 10: any weight/adapter upload API, LoRA fine-tuning support, or documentation of bringing custom checkpoints to Morph's serving layer.
- [claimed-docs] “A small Reflex trains in about 30 seconds.”
- [claimed-docs] “Reserve model capacity by choosing a model and plan. Morph provisions and operates it”
- [claimed-docs] “Point any OpenAI SDK at `https://api.morphllm.com/v1`. One API key covers the open-weight chat models and the specialized tools”
Model catalog — stories about model catalog in this arenaModel catalog
Stories about model catalog in this arena
Catalog
ml-engineerGet newly released open-weight models on the platform quickly after their public release
weight 2 · round drawnTogether AInone0/10The evidence pack shows Together AI hosts 100+ open-source models and supports fine-tuning/dedicated deployment, but there is no evidence—first-party or independent—about the speed or process of adding newly released open-weight models to the catalog. missing for 10: any documentation or community report on time-to-availability for new model releases, changelog/release-notes evidence, or comparison of catalog freshness versus competitors.
Morphnone0/10The evidence shows Morph supports a curated set of open-weight chat models (e.g., Kimi K3) and offers OpenAI/Anthropic-compatible access, but there is no documentation or report about how quickly new open-weight model releases are added to the platform. A community comment even questions why only an older model was listed on OpenRouter, suggesting possible lag, but this is not a concrete contradiction of a specific claim. Missing for 10: any release-cadence documentation, changelog, or independent confirmation of rapid onboarding of newly released open-weight models.
- [claimed-docs] “Move a production agent off Anthropic or OpenAI onto Kimi K3 — a 5% traffic trial or a full migration, gated by Reflexes.”
- [claimed-docs] “Anthropic Messages supports the same open source chat models, token billing, and rate limits.”
- [community] “Last time I looked into Morph, I noticed you weren't yet on OpenRouter. I see that's changed, but it looks like only an older model is liste…”
developerRely on a documented deprecation policy with advance notice before a hosted model is removed
weight 1 · round drawnTogether AInone0/10No evidence in the pack mentions a deprecation policy, sunset schedule, or advance-notice commitment for retiring hosted models; docs cover serverless/dedicated inference, fine-tuning, and agent tooling but nothing about model lifecycle/deprecation notices.
ai-native userHave an agent enumerate the live model catalog programmatically via a documented GET /v1/models-style endpoint
weight 2 · round to Together AITogether AI exposes an OpenAI-compatible /v1/models endpoint confirmed live via probe (returns 401 without key, meaning it exists and speaks JSON, per together-ai-probe-rt-1), and docs confirm OpenAI-compatibility for base URL/API key swaps (together-ai-docs-1) enabling programmatic catalog enumeration. missing for 10: a dedicated docs page explicitly documenting the /v1/models endpoint schema/response format, and independent hands-on confirmation of a successful authenticated call listing the full catalog.
- [probe] “PROBE models-endpoint (2026-09-04): GET https://api.together.xyz/v1/models without a key returned HTTP 401 (Missing API key) — the OpenAI-st…”
- [claimed-docs] “you can point it at models hosted on Together with two changes: the API key and base URL”
Morphnone0/10Morph documents an OpenAI-compatible base URL (/v1) and multiple endpoints (/v1/messages, batch, caching, dedicated endpoints) but no evidence pack item mentions a GET /v1/models catalog endpoint, and the OpenAPI schema probe returned 404s across all candidate paths, meaning no documented machine-readable catalog listing was found.
- [claimed-docs] “Point any OpenAI SDK at `https://api.morphllm.com/v1`. One API key covers the open-weight chat models and the specialized tools”
- [probe] “PROBE openapi: all candidate paths 404 (https://docs.morphllm.com/openapi.json, https://docs.morphllm.com/swagger.json, https://docs.morphll…”
developerChoose among a broad catalog of current open-weight model families (Llama, Qwen, DeepSeek, GPT-OSS and peers) on shared serverless endpoints
weight 3 · round to Together AIDocs confirm serverless endpoints serving '100+ open-source models' with per-token pricing, no provisioning latency, and OpenAI-compatible API for easy model switching, and community evidence corroborates real-world use of Llama-family models on the platform. However, the evidence pack never explicitly enumerates Qwen, DeepSeek, or GPT-OSS by name, so full breadth across all cited model families isn't directly confirmed. Missing for 10: explicit documentation naming Qwen/DeepSeek/GPT-OSS as catalog entries, and independent verification of catalog breadth beyond the generic '100+ models' claim.
- [claimed-docs] “Call 100+ open-source models with per-token pricing and no provisioning latency.”
- [claimed-docs] “you can point it at models hosted on Together with two changes: the API key and base URL”
- [claimed-docs] “you can prototype on serverless and move to reserved hardware later without changing your application code”
- [community] “Colleague and I were discussing need for a service like this. Their pricing is much cheaper than self-hosting Llama-2-70B on AWS ($20.36/hr)…”
- [probe] “PROBE models-endpoint (2026-09-04): GET https://api.together.xyz/v1/models without a key returned HTTP 401 (Missing API key) — the OpenAI-st…”
Morph's docs confirm an OpenAI-compatible endpoint serving 'open-weight chat models' broadly, with prefix caching enabled for 'every open source model' and specific mention of migrating between providers like Kimi K3, but no evidence explicitly lists Llama, Qwen, DeepSeek, GPT-OSS or enumerates a broad multi-family catalog. Missing for 10: an explicit model catalog/list naming specific open-weight families (Llama, Qwen, DeepSeek, GPT-OSS), and independent confirmation of catalog breadth beyond Kimi K3.
- [claimed-docs] “Point any OpenAI SDK at `https://api.morphllm.com/v1`. One API key covers the open-weight chat models and the specialized tools”
- [claimed-docs] “Move a production agent off Anthropic or OpenAI onto Kimi K3 — a 5% traffic trial or a full migration, gated by Reflexes.”
- [claimed-docs] “Prefix caching is on for every open source model. No configuration, no cache-write surcharge.”
- [claimed-docs] “Anthropic Messages supports the same open source chat models, token billing, and rate limits.”
Multimodal — stories about multimodal in this arenaMultimodal
Stories about multimodal in this arena
Modalities
developerGenerate embeddings (and rerank results) for retrieval pipelines without a second vendor
weight 1 · round to Together AIDocs confirm a unified API that includes embeddings alongside chat/image/audio (together-ai-docs-21), supporting retrieval pipelines without a separate embeddings vendor. However, no evidence in the pack documents a dedicated rerank endpoint or reranking models, so the reranking half of the story is unsupported. Missing for 10: dedicated rerank API/model documentation, embeddings model list/dimensions, and independent confirmation of embeddings quality/usage in RAG pipelines.
- [claimed-docs] “Call chat, image, audio, embedding, and more through one API.”
developerCall vision, audio, or image-generation models beyond text chat on the same platform
weight 1 · round to Together AITogether AI's docs explicitly state you can 'Call chat, image, audio, embedding, and more through one API' on the serverless platform, indicating multimodal model support beyond text chat on the same platform [together-ai-docs-21]. This is corroborated by the broader serverless/dedicated inference docs describing a unified API/base URL approach [together-ai-docs-1, together-ai-docs-2]. Missing for 10: no independent hands-on report or example code specifically calling vision/audio/image-generation endpoints, and no detailed model list confirming breadth of non-text modalities.
- [claimed-docs] “Call chat, image, audio, embedding, and more through one API.”
- [claimed-docs] “Call 100+ open-source models with per-token pricing and no provisioning latency.”
- [claimed-docs] “you can point it at models hosted on Together with two changes: the API key and base URL”
Morphnone0/10All evidence describes Morph as a text/code-focused LLM API platform (chat completions, Fast Apply code editing, Reflexes classifiers, batch/caching) with no mention of vision, audio, or image-generation model endpoints. This axis applies to an LLM API platform, but no evidence supports multimodal capability.
Openai compat — stories about openai compat in this arenaOpenai compat
Stories about openai compat in this arena
Compat
ai-native userPlug the provider into coding agents and agent frameworks through documented, first-party integration guides
weight 2 · round to Together AITogether AI documents first-party agent-skills and an MCP server specifically for coding agent integration (docs-agent-skills, docs-12/19/20/24/29/30), confirmed live via probe (probe-4, probe-rt-3), plus OpenAI-compatible base URL swap for framework compatibility (docs-1) and function/tool-calling docs for agentic workflows (docs-8/9). Missing for 10: independent hands-on reports specifically validating the agent-skills/MCP integration with popular frameworks like LangChain or Claude Code.
- [claimed-docs] “you can point it at models hosted on Together with two changes: the API key and base URL”
- [claimed-docs] “Make your AI coding agent Together-AI-aware with ready-made skills for code generation and an MCP server for live docs lookup.”
- [claimed-docs] “12 domain-specific skills that load on demand and teach your agent how to write correct Together AI code (right model IDs, SDK patterns, bes…”
- [claimed-docs] “Gives your agent live access to this documentation site so it can look up current information without leaving your editor.”
- [claimed-docs] “Docs MCP server: Gives your agent live access to this documentation site so it can look up current information without leaving your editor.”
- [claimed-docs] “Function calling (also called tool calling) lets LLMs respond with structured function names and arguments that you can execute in your appl…”
- [claimed-docs] “To build agent loops, chain tool calls inside one response (multi-step), and conversations that thread tools across many turns (multi-turn).”
- [probe] “official MCP server documented at https://docs.together.ai/docs/agent-skills”
- [probe] “PROBE mcp-endpoint (2026-09-04): POST initialize to https://docs.together.ai/mcp answered HTTP 200 with a JSON-RPC/MCP response (event: mess…”
Morph provides first-party documented guides for integrating with coding agents/frameworks: OpenAI-compatible endpoint (morph-docs-1), native Anthropic Messages API for Claude Code (morph-docs-7), MCP server setup for Claude Code (morph-docs-5, morph-docs-9, morph-probe-4), and Codex integration via edit_file (morph-docs-8). Community evidence (goose author, HN discussion) corroborates real-world use in agent frameworks. Missing for 10: independent verification of MCP/agent integration reliability beyond docs and mixed community feedback on apply accuracy.
- [claimed-docs] “Point any OpenAI SDK at `https://api.morphllm.com/v1`. One API key covers the open-weight chat models and the specialized tools”
- [claimed-docs] “Install the Morph MCP so Claude Code can read it”
- [claimed-docs] “Claude Code speaks the Anthropic Messages API, which Morph serves natively at `/v1/messages`.”
- [claimed-docs] “Keep your current Codex model and give it `edit_file` backed by `morph-v3-fast` at 10,500+ tok/s”
- [claimed-docs] “Connect your favorite AI tools to Morph's blazing-fast file editing via Model Context Protocol.”
- [probe] “official MCP server documented at https://docs.morphllm.com/guides/mcp”
- [community] “Have been using morph for a while (I am one of the authors of goose) and was surprised when introduced at the boost it gave me (much less it…”
ai-native userHave an agent switch to or away from this provider mid-workflow because it speaks the standard chat-completions API without provider-specific code changes
weight 3 · round drawnDocs explicitly state OpenAI-compatibility requires only changing API key and base URL, meaning agents/tools written against the standard chat-completions API can point at Together with no provider-specific code changes; the live /v1/models endpoint confirms OpenAI-style REST shape. This directly supports switching to/away from Together mid-workflow with minimal friction. Missing for 10: independent third-party confirmation of drop-in swap behavior (e.g., a framework's changelog showing zero-code-change provider swap) and explicit mention of streaming/tool-call parity with OpenAI's spec beyond docs.
- [claimed-docs] “you can point it at models hosted on Together with two changes: the API key and base URL”
- [claimed-docs] “you can prototype on serverless and move to reserved hardware later without changing your application code”
- [probe] “PROBE models-endpoint (2026-09-04): GET https://api.together.xyz/v1/models without a key returned HTTP 401 (Missing API key) — the OpenAI-st…”
Morph explicitly documents an OpenAI-compatible endpoint (and Anthropic Messages compatibility) plus first-class provider-switching workflows—canary traffic splits and full migration from OpenAI/Anthropic to another model—showing the standard API enables mid-workflow provider switches without code changes. Community evidence (morph-comm-5) independently confirms the OpenAI-compatible endpoint works with third-party tools. missing for 10: independent hands-on verification of the canary/migration switching feature itself (only vendor docs), and no evidence of switching away from Morph to another provider.
- [claimed-docs] “Point any OpenAI SDK at `https://api.morphllm.com/v1`. One API key covers the open-weight chat models and the specialized tools”
- [claimed-docs] “Move a production agent off Anthropic or OpenAI onto Kimi K3 — a 5% traffic trial or a full migration, gated by Reflexes.”
- [claimed-docs] “A canary keeps your current provider, sends a slice of production traffic to Kimi K3, and compares Reflex results before you switch.”
- [claimed-docs] “Anthropic Messages supports the same open source chat models, token billing, and rate limits.”
- [probe] “PROBE llms.txt: HTTP 200 at https://docs.morphllm.com/llms.txt # Morph API Documentation > https://docs.morphllm.com > Base URL: https://api…”
- [community] “This uses an OpenAI-compatible endpoint, so got this working with my llm.datasette.io CLI tool... One thing that worries me: since it's usin…”
developerPoint an existing OpenAI SDK client at the provider by changing only the base URL and API key
weight 3 · round drawnDocs explicitly state you can point an existing client at Together by changing only the API key and base URL, and the API is OpenAI-compatible with a live /v1/models endpoint confirmed via probe. missing for 10: independent developer testimonial specifically confirming drop-in OpenAI SDK compatibility (community evidence covers pricing/UX but not this specific mechanic), and no explicit code snippet shown in evidence.
- [claimed-docs] “you can point it at models hosted on Together with two changes: the API key and base URL”
- [probe] “PROBE models-endpoint (2026-09-04): GET https://api.together.xyz/v1/models without a key returned HTTP 401 (Missing API key) — the OpenAI-st…”
- [claimed-docs] “you can prototype on serverless and move to reserved hardware later without changing your application code”
Docs explicitly state to point any OpenAI SDK at https://api.morphllm.com/v1 with a single API key, corroborated by a probe confirming the llms.txt describes it as OpenAI-compatible, and community evidence confirms it was actually wired up with an OpenAI-compatible client tool. missing for 10: independent hands-on confirmation of a truly drop-in swap with zero code changes beyond base URL/key, and no official OpenAPI spec found to verify full compatibility surface.
- [claimed-docs] “Point any OpenAI SDK at `https://api.morphllm.com/v1`. One API key covers the open-weight chat models and the specialized tools”
- [probe] “PROBE llms.txt: HTTP 200 at https://docs.morphllm.com/llms.txt # Morph API Documentation > https://docs.morphllm.com > Base URL: https://api…”
- [community] “This uses an OpenAI-compatible endpoint, so got this working with my llm.datasette.io CLI tool... One thing that worries me: since it's usin…”
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 to Together AIDocs show broad API/CLI/SDK coverage mirroring UI actions—fine-tuning can be launched 'from the console, through the API/SDK, or with the CLI', dedicated endpoints can be deployed via CLI (tg beta endpoints deploy), and inference, batch, and cluster provisioning are all API-accessible—suggesting strong UI/API parity for core workflows. However, there's no explicit documentation enumerating full parity (e.g., billing, account/org management, usage dashboards) or confirming every UI-only feature has an API equivalent, and no OpenAPI spec was found (probe shows all openapi.json paths 404), making completeness hard to verify. Missing for 10: an explicit parity statement or API coverage for account/billing/admin functions, and a published OpenAPI spec confirming full API surface.
- [claimed-docs] “You can launch a fine-tuning job from the console, through the API/SDK, or with the CLI”
- [claimed-docs] “tg beta endpoints deploy google/gemma-4-E4B-it --endpoint my-endpoint”
- [claimed-docs] “Together AI handles the full lifecycle: data upload, training, hosting, and inference on a dedicated endpoint.”
- [claimed-docs] “Spin up H100 and B200 clusters with attached storage for training or large batch jobs.”
- [claimed-docs] “Run asynchronous batch workloads at up to 50% lower cost.”
- [probe] “PROBE openapi: all candidate paths 404 (https://docs.together.ai/openapi.json, https://docs.together.ai/swagger.json, https://docs.together.…”
Morph is fundamentally API-first — nearly all documented functionality (chat completions, Fast Apply, Compact, Reflexes, batch, dedicated endpoints, MCP) is exposed via the API/SDK per morph-docs-1 through morph-docs-16, and a playground/dashboard UI is only mentioned in passing (morph-comm-2). There's no explicit documentation enumerating dashboard-only features (billing, endpoint provisioning UI, playground testing) and confirming each has an API equivalent, so full parity isn't demonstrated. Missing for 10: explicit UI/dashboard feature list mapped to API endpoints, confirmation that account/billing/endpoint-management actions available in the dashboard are also API-accessible, and independent verification of parity.
- [claimed-docs] “Point any OpenAI SDK at `https://api.morphllm.com/v1`. One API key covers the open-weight chat models and the specialized tools”
- [claimed-docs] “Reserve model capacity by choosing a model and plan. Morph provisions and operates it”
- [community] “Seems completely broken. I used the provided HTML example on morphllm.com playground. Without editing anything, I pressed apply. Your model …”
- [probe] “PROBE llms.txt: HTTP 200 at https://docs.morphllm.com/llms.txt # Morph API Documentation > https://docs.morphllm.com > Base URL: https://api…”
ai-native userExport all of my data in open formats and leave
weight 3 · round drawnTogether AInone0/10The evidence pack covers inference, fine-tuning, dedicated endpoints, and agent tooling, but contains no mention of a data export feature, downloadable account data, or a documented way to retrieve fine-tuning datasets/model weights in open formats and leave the platform. This is a fair question for a hosted AI platform that stores prompts, fine-tuning data, and trained models, so absence of evidence means 'none' rather than 'na'.
ai-native userRead the product's source under an open license
weight 2 · round drawnTogether AInone0/10The evidence pack covers Together AI's hosted inference, fine-tuning, and dedicated endpoint features, and its client SDKs (e.g., together-typescript) are open on GitHub, but there is no evidence that Together AI's core platform/product source is released under an open license—only that it serves open-source models to customers.
Pricing limits — free-tier ceilings, usage caps, and rate limits before you have to payPricing limits
Free-tier ceilings, usage caps, and rate limits before you have to pay
Limits
developerRead documented rate limits and how they scale across usage tiers before I hit them in production
weight 2 · round to Together AITogether AI documents a rate-limit model explicitly (docs-14): dynamic per-organization, per-model limits based on live model capacity and recent successful usage, and notes dedicated endpoints have no hard rate limits (docs-31). This gives developers some visibility into how limits work before hitting them, but there's no documented tiered structure (e.g., spend-based tiers with specific numeric thresholds) showing how limits scale as usage/spend grows over time. Missing for 10: explicit usage-tier table or thresholds, numeric rate-limit values per tier, guidance on how to request/track tier upgrades.
- [claimed-docs] “Together uses dynamic rate limits instead of fixed thresholds. Each organization has a dynamic rate per model that adjusts based on: The mod…”
- [claimed-docs] “Dedicated model inference (DMI) lets you serve a model on reserved hardware... Better performance... No hard rate limits”
Morphnone0/10The evidence mentions rate limits exist (morph-docs-16: 'same open source chat models, token billing, and rate limits') but never documents actual limit numbers, tiers, or how they scale with usage — no pricing/tier table or rate-limit reference is shown. Missing for 10: an actual rate-limit table or tier breakdown, numeric thresholds, and documentation on how limits change across plans.
- [claimed-docs] “Anthropic Messages supports the same open source chat models, token billing, and rate limits.”
- [claimed-docs] “Reserve model capacity by choosing a model and plan. Morph provisions and operates it”
Pricing
founderSet spending caps or budget alerts so a runaway workload cannot generate an unbounded bill
weight 1 · round drawnTogether AInone0/10No evidence of spending caps, budget alerts, or usage-based cutoffs; docs mention rate limits (dynamic, based on capacity/usage) and pricing structures but nothing about setting a spend ceiling or receiving alerts to prevent runaway bills.
Morphnone0/10No evidence pack item mentions spending caps, budget alerts, usage limits, or billing controls of any kind; docs focus on models, tools, and pricing mechanics like batch/cache but not cap-setting features. This axis applies to any paid API/cloud product but no capability is documented, so it's 'none' rather than 'na'.
founderSee public per-token prices for every hosted model without talking to sales
weight 3 · round to Together AIDocs advertise "per-token pricing" for 100+ serverless models and community evidence confirms self-serve signup with visible per-token cost (e.g. "$0.0001/1K tokens", $25 trial credit) with no sales contact required. However, no evidence pack item directly shows a public pricing page enumerating rates for every hosted model, and the live models API requires an API key to enumerate the catalog. Missing for 10: a documented public pricing page/table covering all models, independent confirmation that all model prices (not just examples) are visible pre-signup.
- [claimed-docs] “Call 100+ open-source models with per-token pricing and no provisioning latency.”
- [community] “the price is pretty low... 4B MODEL, PRICE 1K TOKENS: $0.0001. register with an email, test account has $25 credit, python API as well, good…”
- [community] “I've explored most of the inference as a service platforms, theirs is particularly nice. (Clear/simple UI, good pricing, good speed.)”
- [probe] “PROBE models-endpoint (2026-09-04): GET https://api.together.xyz/v1/models without a key returned HTTP 401 (Missing API key) — the OpenAI-st…”
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 drawnTogether AInone0/10No evidence in the pack mentions data residency, regional storage options, or geographic control over where data/models are hosted; only privacy policy criticism unrelated to residency is present. Missing for 10: any documentation of region selection, data residency guarantees, or geo-specific hosting options.
Morphnone0/10No evidence in the pack mentions data residency, regional storage options, or geographic controls; the closest is a comment about retention/training policy tiers (morph-comm-6), which is about usage not location. Missing for 10: any documentation of region selection, data residency guarantees, or geo-specific hosting options.
- [community] “Please clarify your privacy policy. If my interpretation is correct, paying users will have their data retained and trained on? ... Free Tie…”
ai-native userPrevent my data from being used to train AI models
weight 3 · round to MorphTogether AInone0/10No first-party documentation describes an opt-out, data-retention control, or 'no training on your data' policy; the only relevant evidence is a community report describing Together AI's privacy policy as broad/vague, allowing data use 'for other purposes,' which points toward the opposite of a training-opt-out guarantee.
- [community] “Together AI's privacy policy collects broad personal data including prompts, IP addresses, and contact info, with vague language allowing us…”
- [community] “Colleague and I were discussing need for a service like this. Their pricing is much cheaper than self-hosting Llama-2-70B on AWS ($20.36/hr)…”
A community-sourced excerpt of Morph's privacy policy (morph-comm-6) shows Enterprise Tier users can opt out of having code/data used for training, satisfying the story, but Free Tier data is explicitly used to train models by default — a significant caveat requiring a paid tier to actually prevent training use. No first-party Morph docs in the evidence pack directly address data retention/training opt-out. Missing for 10: first-party documentation of the privacy/data-use policy and opt-out mechanism, independent confirmation that Enterprise-tier exclusion is honored in practice.
- [community] “Please clarify your privacy policy. If my interpretation is correct, paying users will have their data retained and trained on? ... Free Tie…”
ai-native userControl data retention and deletion
weight 2 · round drawnTogether AInone0/10No evidence pack item documents user-facing data retention or deletion controls (e.g., API/dashboard settings to delete stored prompts or data). The only relevant community evidence (together-ai-comm-1) actually raises concerns that Together AI's privacy policy allows broad data collection with vague retention/usage language, which is the opposite of demonstrating retention/deletion controls.
- [community] “Together AI's privacy policy collects broad personal data including prompts, IP addresses, and contact info, with vague language allowing us…”
Morphnone0/10Evidence only shows a general privacy-policy statement about free vs enterprise data usage (morph-comm-6), with a community member explicitly asking for clarification — there is no documented user-facing control for data retention settings or deletion requests. Missing for 10: any docs on data retention windows, user-initiated deletion/export tools, or enterprise data handling guarantees beyond the raw policy snippet.
- [community] “Please clarify your privacy policy. If my interpretation is correct, paying users will have their data retained and trained on? ... Free Tie…”
ai-native userOpt out of telemetry and usage tracking
weight 2 · round to MorphTogether AInone0/10No documentation or feature reference describes a telemetry/usage-tracking opt-out mechanism; the only related evidence is a community critique noting Together AI's privacy policy broadly collects prompts, IPs, and contact info with vague 'other purposes' language and no opt-out mentioned, reinforcing the absence of this capability.
- [community] “Together AI's privacy policy collects broad personal data including prompts, IP addresses, and contact info, with vague language allowing us…”
- [community] “Colleague and I were discussing need for a service like this. Their pricing is much cheaper than self-hosting Llama-2-70B on AWS ($20.36/hr)…”
Community evidence indicates a tiered privacy policy where free-tier code may be used for training while Enterprise-tier data is not used beyond immediate processing, implying paying/enterprise users can effectively opt out of data usage — but there is no explicit telemetry/usage-tracking opt-out control or documentation for individual users. missing for 10: explicit telemetry opt-out toggle/setting, first-party privacy documentation confirming opt-out mechanics, independent confirmation beyond a single HN comment.
- [community] “Please clarify your privacy policy. If my interpretation is correct, paying users will have their data retained and trained on? ... Free Tie…”
Reliability status — stories about reliability status in this arenaReliability status
Stories about reliability status in this arena
Reliability
founderGet a stated availability SLA on paid or enterprise tiers
weight 1 · round drawnTogether AInone0/10Evidence shows a public status page and various product docs, but nothing states an actual uptime/availability SLA percentage or enterprise contractual guarantee. Absence of evidence for this applicable capability yields none.
- [probe] “PROBE status-page (2026-09-04): https://status.together.ai returns HTTP 200 and renders a public service-status page (page body includes "op…”
Morphnone0/10No evidence pack item mentions an SLA, uptime commitment, or reliability guarantee for paid/enterprise tiers — dedicated endpoints and enterprise privacy terms are mentioned but no stated availability percentage or SLA terms are documented.
- [claimed-docs] “Reserve model capacity by choosing a model and plan. Morph provisions and operates it”
- [community] “Please clarify your privacy policy. If my interpretation is correct, paying users will have their data retained and trained on? ... Free Tie…”
founderCheck a public status page with incident history before betting production traffic on the platform
weight 2 · round to Together AIA live public status page at status.together.ai returns HTTP 200 and shows service status ('operational'), confirming a real status dashboard exists that a founder could check before committing production traffic. Missing for 10: no evidence of detailed incident history/postmortems or uptime SLA history visible in the pack, so depth of incident history can't be fully confirmed.
- [probe] “PROBE status-page (2026-09-04): https://status.together.ai returns HTTP 200 and renders a public service-status page (page body includes "op…”
Speed latency — stories about speed latency in this arenaSpeed latency
Stories about speed latency in this arena
Benchmarks
ml-engineerSee published tokens-per-second or latency numbers, benchmarks, or load-testing guides backing the provider's speed claims
weight 2 · round to MorphTogether AInone0/10The evidence pack contains only vague qualitative claims ('no provisioning latency', 'Better performance' for dedicated endpoints) and subjective community remarks ('good speed') but no published tokens-per-second figures, latency benchmarks, or load-testing methodology/guides that would let an ml-engineer verify speed claims.
- [claimed-docs] “Call 100+ open-source models with per-token pricing and no provisioning latency.”
- [claimed-docs] “Dedicated model inference (DMI) lets you serve a model on reserved hardware... Better performance... No hard rate limits”
- [community] “I've explored most of the inference as a service platforms, theirs is particularly nice. (Clear/simple UI, good pricing, good speed.)”
Morph publishes specific throughput numbers (10,500+ tok/s Fast Apply, 33,000 tok/s Compact, 90ms classifiers, 30s Reflex training) across its docs and even one independent community post corroborates a speed/iteration boost, but there is no independent benchmark suite, third-party load-testing report, or reproducible methodology, and one community report describes inaccurate output rather than confirming throughput. missing for 10: independent benchmark/load-testing reports, published methodology or test conditions behind the tok/s figures, and resolution of the community accuracy complaint.
- [claimed-docs] “Keep your current Codex model and give it `edit_file` backed by `morph-v3-fast` at 10,500+ tok/s”
- [claimed-docs] “Fast Apply merges just the changed lines at 10,500 tok/s, 98% accuracy.”
- [claimed-docs] “Drop filler from chat history and code context at 33,000 tok/s. 50-70% reduction, every surviving line byte-for-byte identical to input.”
- [claimed-docs] “Small text classifiers that label a turn in ~90ms — jailbreaks, NSFW, stuck-in-a-loop, user frustration.”
- [claimed-docs] “A small Reflex trains in about 30 seconds.”
- [github] “Use Morph's API to intelligently modify existing code at 4,500+ tokens/sec”
- [community] “Have been using morph for a while (I am one of the authors of goose) and was surprised when introduced at the boost it gave me (much less it…”
- [community] “Seems completely broken. I used the provided HTML example on morphllm.com playground. Without editing anything, I pressed apply. Your model …”
Serving
developerServe latency-sensitive workloads with fast time-to-first-token and high-throughput generation
weight 3 · round to MorphTogether AIdisputedcontradicted5/10Together AI's docs claim strong speed characteristics — 'no provisioning latency' for serverless, DMI offering 'better performance' and 'no hard rate limits', dynamic rate limits based on live capacity, streaming SSE support, and automatic prompt caching to speed repeated prefills — all directly supporting fast TTFT and high throughput [together-ai-docs-2][together-ai-docs-31][together-ai-docs-14][together-ai-gh-1][together-ai-docs-13]. However, an independent hands-on comparison explicitly disputes this, stating Together AI was 'at the bottom of the pile' for large-scale deployment speed/pricing versus competitors like Cerebras, DeepInfra, and Hyperbolic [together-ai-comm-5], while another community post gives a positive but vaguer 'good speed' impression [together-ai-comm-4] — a genuine mixed/contradicted signal rather than mere skepticism. Missing for 10: independent benchmarks (TTFT/tokens-per-second numbers) confirming or refuting the vendor's latency claims, and resolution of the conflicting community reports.
- [claimed-docs] “Call 100+ open-source models with per-token pricing and no provisioning latency.”
- [claimed-docs] “Dedicated model inference (DMI) lets you serve a model on reserved hardware... Better performance... No hard rate limits”
- [claimed-docs] “Together uses dynamic rate limits instead of fixed thresholds. Each organization has a dynamic rate per model that adjusts based on: The mod…”
- [claimed-docs] “Automatic: There is no header, parameter, or account toggle to enable it. Send the same prompt prefix again and any portion that's still war…”
- [github] “We provide support for streaming responses using Server Sent Events (SSE).”
- [community] “I've explored most of the inference as a service platforms, theirs is particularly nice. (Clear/simple UI, good pricing, good speed.)”
- [community] “'Together AI has quickly grown into the leading hosting platform for AI' - Lol. Definitely not. I went through pricing and large scale deplo…”
Morph publishes concrete high-throughput numbers (10,500+ tok/s Fast Apply, 33,000 tok/s Compact, ~90ms classifiers, prefix caching with no surcharge) aimed squarely at low-latency/high-throughput serving, and independent community use (goose author) corroborates a real speed boost in practice. Missing for 10: explicit time-to-first-token benchmarks and independent third-party throughput verification beyond vendor docs and one anecdotal HN comment.
- [claimed-docs] “Keep your current Codex model and give it `edit_file` backed by `morph-v3-fast` at 10,500+ tok/s”
- [claimed-docs] “Fast Apply merges just the changed lines at 10,500 tok/s, 98% accuracy.”
- [claimed-docs] “Drop filler from chat history and code context at 33,000 tok/s. 50-70% reduction, every surviving line byte-for-byte identical to input.”
- [claimed-docs] “Small text classifiers that label a turn in ~90ms — jailbreaks, NSFW, stuck-in-a-loop, user frustration.”
- [claimed-docs] “Prefix caching is on for every open source model. No configuration, no cache-write surcharge.”
- [community] “Have been using morph for a while (I am one of the authors of goose) and was surprised when introduced at the boost it gave me (much less it…”
ml-engineerBenefit from prompt/prefix caching that reduces latency or cost on repeated context
weight 1 · round drawnTogether AI documents automatic prompt/prefix caching for serverless chat models, explicitly stating cached input tokens are billed at a steep discount with no toggle needed, directly reducing cost on repeated context. missing for 10: independent/hands-on benchmark confirming latency reduction, and clarity on which models/latency gains apply beyond the cost-discount framing.
- [claimed-docs] “Automatic: There is no header, parameter, or account toggle to enable it. Send the same prompt prefix again and any portion that's still war…”
- [claimed-docs] “Send the same prompt prefix again and any portion that's still warm in the shared cache is billed at the cached rate.”
- [claimed-docs] “Select serverless chat models bill cached input tokens at a steep discount... Automatic: There is no header, parameter, or account toggle to…”
Morph docs explicitly state prefix caching is enabled by default for all open-source models with no configuration and no cache-write surcharge, directly addressing latency/cost reduction on repeated context. Missing for 10: independent benchmark or community corroboration of actual latency/cost savings from caching specifically (community evidence covers Fast Apply speed, not caching).
- [claimed-docs] “Prefix caching is on for every open source model. No configuration, no cache-write surcharge.”
developerStream completions token by token over SSE for responsive user experiences
weight 3 · round to Together AITogether AI's SDKs explicitly document SSE-based streaming support ('We provide support for streaming responses using Server Sent Events (SSE)'), and the API is OpenAI-compatible, implying standard streaming chat completion endpoints. Missing for 10: no independent hands-on latency benchmark or explicit token-by-token example snippet in the evidence pack.
- [github] “We provide support for streaming responses using Server Sent Events (SSE).”
- [github] “This library provides convenient access to the Together REST API from server-side TypeScript or JavaScript.”
- [claimed-docs] “you can point it at models hosted on Together with two changes: the API key and base URL”
Morph advertises full OpenAI SDK/API compatibility at api.morphllm.com/v1 (morph-docs-1), which implies standard streaming support, but no evidence explicitly documents SSE/token-by-token streaming, a `stream` parameter, or any latency/responsiveness benchmarks for streamed output. missing for 10: explicit docs or examples showing `stream=true`/SSE usage, first-party or community confirmation of token-by-token streaming behavior, and any measured time-to-first-token metrics.
- [claimed-docs] “Point any OpenAI SDK at `https://api.morphllm.com/v1`. One API key covers the open-weight chat models and the specialized tools”
- [probe] “PROBE llms.txt: HTTP 200 at https://docs.morphllm.com/llms.txt # Morph API Documentation > https://docs.morphllm.com > Base URL: https://api…”
Structured tool calling — stories about structured tool calling in this arenaStructured tool calling
Stories about structured tool calling in this arena
Structured
developerEnforce structured outputs against a JSON schema (or grammar) so model responses parse reliably
weight 3 · round to Together AITogether AI's docs explicitly describe structured outputs where supported models return JSON conforming to a supplied schema, avoiding retries/fragile parsing, alongside function/tool calling and multi-step agentic tool call chaining. This directly matches the story's request for JSON-schema-enforced structured outputs; missing for 10: no independent/hands-on validation of schema-conformance reliability or grammar-based constraints beyond JSON schema.
- [claimed-docs] “Supported models can return JSON that conforms to any schema you supply, so you can read the output directly in code without retries or frag…”
- [claimed-docs] “Function calling (also called tool calling) lets LLMs respond with structured function names and arguments that you can execute in your appl…”
- [claimed-docs] “To build agent loops, chain tool calls inside one response (multi-step), and conversations that thread tools across many turns (multi-turn).”
- [claimed-docs] “Multi-step function calling chains sequential function calls within one conversation turn.”
Morphnone0/10Morph's docs cover OpenAI-compatible endpoints, Fast Apply, compaction, batch, caching, and MCP integration, but no evidence describes JSON schema/grammar-constrained structured output enforcement for reliable parsing. Missing for 10: any mention of response_format/json_schema support, grammar constraints, or validation guarantees on model outputs.
Tools
ai-native userRely on faithful function/tool calling — including parallel and multi-step tool use — so agent loops run on open models without breaking
weight 3 · round to Together AITogether AI documents native function/tool calling with explicit support for multi-step (chained) and multi-turn agentic tool use, positioned as the foundation for agent loops (together-ai-docs-8,9,15,16). However, there is no independent/hands-on evidence confirming reliability of parallel tool calls or that it avoids breakage across open models in practice — only first-party docs. missing for 10: independent benchmarks or community reports on parallel tool-call accuracy, explicit parallel tool-calling documentation, and hands-on verification that agent loops don't break on open models.
- [claimed-docs] “Function calling (also called tool calling) lets LLMs respond with structured function names and arguments that you can execute in your appl…”
- [claimed-docs] “To build agent loops, chain tool calls inside one response (multi-step), and conversations that thread tools across many turns (multi-turn).”
- [claimed-docs] “Function calling (also called *tool calling*) lets LLMs respond with structured function names and arguments that you can execute in your ap…”
- [claimed-docs] “Multi-step function calling chains sequential function calls within one conversation turn.”
Morph documents OpenAI-compatible and Anthropic Messages-compatible endpoints and specialized tools like edit_file/Fast Apply meant to plug into agent loops (Claude Code, Codex) with high speed and claimed 98% accuracy, which implies decent tool-call fidelity for open models. However there's no explicit documentation of parallel or multi-step tool-call handling, and a hands-on community report describes the apply/tool behavior as 'completely broken' with unrequested edits, undercutting full confidence in reliability. Missing for 10: explicit parallel tool-call support docs, multi-step agent-loop stability guarantees, and independent verification resolving the reported apply failures.
- [claimed-docs] “Claude Code speaks the Anthropic Messages API, which Morph serves natively at `/v1/messages`.”
- [claimed-docs] “Keep your current Codex model and give it `edit_file` backed by `morph-v3-fast` at 10,500+ tok/s”
- [claimed-docs] “Fast Apply merges just the changed lines at 10,500 tok/s, 98% accuracy.”
- [claimed-docs] “Connect your favorite AI tools to Morph's blazing-fast file editing via Model Context Protocol.”
- [community] “Seems completely broken. I used the provided HTML example on morphllm.com playground. Without editing anything, I pressed apply. Your model …”
- [community] “Does Claude Code have a similar apply model? It does create diffs for you to accept/reject but then I feel like it's always using a find/rep…”
Not comparable on these axes
ai-native userPlug MCP servers into this product so it can use their tools
weight 3 · not comparableTogether AInone0/10The evidence only shows Together AI publishing its own MCP server (docs lookup) for coding agents to consume — i.e., Together acting as an MCP *server*, not as a client that lets users plug external MCP servers into it. There's no documentation or probe showing Together's platform/API can connect to and use tools from user-supplied MCP servers, despite having function/tool-calling APIs. missing for 10: any evidence of MCP-client support (configuring/connecting external MCP servers within Together's product), integration of MCP tool results into its agentic function-calling flow.
- [claimed-docs] “Make your AI coding agent Together-AI-aware with ready-made skills for code generation and an MCP server for live docs lookup.”
- [claimed-docs] “12 domain-specific skills that load on demand and teach your agent how to write correct Together AI code (right model IDs, SDK patterns, bes…”
- [claimed-docs] “Gives your agent live access to this documentation site so it can look up current information without leaving your editor.”
- [claimed-docs] “Docs MCP server: Gives your agent live access to this documentation site so it can look up current information without leaving your editor.”
- [probe] “official MCP server documented at https://docs.together.ai/docs/agent-skills”
- [probe] “PROBE mcp-endpoint (2026-09-04): POST initialize to https://docs.together.ai/mcp answered HTTP 200 with a JSON-RPC/MCP response (event: mess…”
- [claimed-docs] “To build agent loops, chain tool calls inside one response (multi-step), and conversations that thread tools across many turns (multi-turn).”
- [claimed-docs] “Multi-step function calling chains sequential function calls within one conversation turn.”
Morphn/aMorph is an LLM API/tool-serving platform (fast-apply, reflexes, model routing) that itself exposes an MCP server for other agents (e.g., Claude Code) to consume — it is not an agent host that ingests external MCP servers as tool sources. The story's client-side capability (plugging MCP servers into Morph so it can use their tools) is a category mismatch for this product type; all MCP evidence describes Morph being served TO clients, not consuming servers itself.
- [claimed-docs] “Install the Morph MCP so Claude Code can read it”
- [claimed-docs] “Connect your favorite AI tools to Morph's blazing-fast file editing via Model Context Protocol.”
- [probe] “official MCP server documented at https://docs.morphllm.com/guides/mcp”
ai-native userGet AI-generated insights and suggestions from my data inside the product
weight 2 · not comparableTogether AIn/aTogether AI is an inference/training infrastructure API platform; it does not host 'your data' inside a product surface with dashboards or analytics from which it generates insights and suggestions — that role belongs to end-user applications built on top of it, not to Together AI itself.
ai-native userDelegate tasks to a built-in AI assistant inside the product
weight 3 · not comparableTogether AInone0/10The axis applies to this product kind (peer products hold positive or none verdicts on this story), so lack of evidence for an applicable capability is "none", never "na". (na/none harmonized at arena bring-up — see pipeline/scripts/na-harmonize.ts.)
Morphn/aMorph is a developer-facing API/infrastructure platform (fast-apply, chat models, MCP tools) meant to be embedded into other products or coding agents — it is not itself an end-user product with a built-in assistant UI to delegate tasks to. This axis is a category error for an API/infra offering, not a missing feature.
ai-native userOperate the product with natural-language commands
weight 2 · not comparableTogether AInone0/10Together AI is an inference/training API platform operated via REST API, SDKs, CLI, and console — the evidence shows structured commands (API calls, CLI syntax like 'tg beta endpoints deploy...') rather than any natural-language command interface for operating the platform itself. The agent-skills/MCP docs only help external coding agents write correct Together AI code, not let a user operate Together AI's own product surface via natural language.
Morphn/aMorph is a backend model-serving/API and MCP-tool provider consumed programmatically by coding agents (Claude Code, Cursor, etc.); it is not itself a user-facing surface where an end user types natural-language commands. The natural-language interface lives in the calling agent, not in Morph, so this axis is a category mismatch for this kind of infrastructure product.
ai-native userTest against a sandbox environment without touching production data
weight 1 · not comparableTogether AInone0/10No evidence of a dedicated sandbox/staging environment; the only related mention is a test account with trial credit on the same production API, not an isolated sandbox for testing without touching production data.
- [community] “the price is pretty low... 4B MODEL, PRICE 1K TOKENS: $0.0001. register with an email, test account has $25 credit, python API as well, good…”
ai-native userDefine rules that trigger actions automatically on events
weight 3 · not comparableTogether AIn/aTogether AI is an inference/training/hosting API platform, not a workflow-automation or event-rule system; there is no evidence of a rules engine, triggers, or event-driven actions being a category feature. This story targets automation-platform capabilities that fall outside Together AI's product scope (model inference, fine-tuning, dedicated endpoints), so the axis does not apply.
Morph's 'Reflexes' are classifiers that detect events (jailbreak, frustration, loop) and can gate automatic actions like canary traffic-splitting or full migration between models, which is a narrow form of event-triggered automation. However there's no documented general-purpose rule engine letting users define arbitrary custom rules/actions beyond this migration-gating use case. Missing for 10: a general user-facing rules/automation builder, documentation of broader trigger-action mappings beyond model migration, and independent confirmation of custom Reflex-to-action wiring.
- [claimed-docs] “Move a production agent off Anthropic or OpenAI onto Kimi K3 — a 5% traffic trial or a full migration, gated by Reflexes.”
- [claimed-docs] “A canary keeps your current provider, sends a slice of production traffic to Kimi K3, and compares Reflex results before you switch.”
- [claimed-docs] “A small Reflex trains in about 30 seconds.”
ai-native userVersion, review, and roll back my automations
weight 1 · not comparableTogether AIn/aTogether AI is an inference/training API platform, not an automation/workflow builder; there is no concept of user-created 'automations' to version, review, or roll back in this evidence pack.
Morphnone0/10Morph is an API/tooling platform for fast code editing, model routing, and reflex classifiers — there's no evidence of any versioning, review, or rollback mechanism for 'automations' (e.g., no workflow history, diff review UI for automation configs, or revert capability). Nothing in the evidence pack addresses this axis at all.
ai-native userSelf-host the core product
weight 3 · not comparableTogether AIn/aTogether AI is a hosted inference/fine-tuning cloud service, not open-source software distributed for self-hosting; nothing in the evidence describes deploying the core Together platform on one's own infrastructure. Self-hosting is a category error for this SaaS-style product.
Morphnone0/10Morph is presented entirely as a hosted API/SaaS product (api.morphllm.com endpoints, dedicated endpoints as reserved capacity, prefix caching, batch processing) with no mention of on-premises deployment, downloadable server binaries, Docker images, or open-source release of the core service. Evidence pack contains no self-hosting instructions or claims.
- [claimed-docs] “Point any OpenAI SDK at `https://api.morphllm.com/v1`. One API key covers the open-weight chat models and the specialized tools”
- [claimed-docs] “Reserve model capacity by choosing a model and plan. Morph provisions and operates it”
- [probe] “PROBE llms.txt: HTTP 200 at https://docs.morphllm.com/llms.txt # Morph API Documentation > https://docs.morphllm.com > Base URL: https://api…”
- [probe] “PROBE openapi: all candidate paths 404 (https://docs.morphllm.com/openapi.json, https://docs.morphllm.com/swagger.json, https://docs.morphll…”