Sakana Marlin vs Gemini Notebook (NotebookLM)

Agent access

1. ai-native userRun the product headlessly / in CI for automation

   weight 2 · round drawn
   Sakana Marlin—none0/10

   Marlin is presented as a research-report generation web product with a UI and pay-per-use credits, with no CLI, API, SDK, or webhook documentation for headless/CI usage; probes for llms.txt and OpenAPI specs both returned 404, indicating no programmatic interface is exposed.

   • [probe] “PROBE llms.txt: HTTP 404 at https://sakana.ai/llms.txt”
   • [probe] “PROBE openapi: all candidate paths 404 (https://sakana.ai/openapi.json, https://sakana.ai/swagger.json, https://sakana.ai/api/openapi.json, …”
   • [claimed-docs] “Add your card. Start right away.”

   proof ↗⚑ flag

   Gemini Notebook (NotebookLM)—none0/10

   No evidence of an API, CLI, or headless/automation mode for NotebookLM/Gemini Notebook; all documented usage is through the interactive web/app UI, and community comments even complain about lacking programmatic source ingestion.

   • [claimed-docs] “Easily upload PDFs, websites, YouTube videos, audio files, Google Docs, or Google Slides or discover new sources.”
   • [community] “My problem with NotebookLM was that you had to give it URLs or docs, you couldn't just have it search for docs and then incorporate them int…”

   proof ↗⚑ flag

2. ai-native userUse an official CLI

   weight 2 · round drawn
   Sakana Marlin—none0/10

   No evidence of an official CLI; Marlin appears to be a web-based research tool with a pay-per-credit UI, and probes for API/llms.txt endpoints returned 404s, suggesting no developer-facing interface is exposed.

   • [probe] “PROBE llms.txt: HTTP 404 at https://sakana.ai/llms.txt”
   • [probe] “PROBE openapi: all candidate paths 404 (https://sakana.ai/openapi.json, https://sakana.ai/swagger.json, https://sakana.ai/api/openapi.json, …”

   proof ↗⚑ flag

   Gemini Notebook (NotebookLM)—none0/10

   No evidence of an official CLI for Gemini Notebook; all documentation describes a web/app UI with no command-line interface mentioned.

   ⚑ flag

3. ai-native userDrive the product through a documented public API

   weight 3 · round drawn
   Sakana Marlin—none0/10

   No documented public API is evidenced; probes for llms.txt and OpenAPI/swagger endpoints all returned 404s, and all other evidence describes the product's research capabilities, not a programmatic interface.

   • [probe] “PROBE llms.txt: HTTP 404 at https://sakana.ai/llms.txt”
   • [probe] “PROBE openapi: all candidate paths 404 (https://sakana.ai/openapi.json, https://sakana.ai/swagger.json, https://sakana.ai/api/openapi.json, …”

   proof ↗⚑ flag

   Gemini Notebook (NotebookLM)—none0/10

   No evidence pack item mentions a documented public API, SDK, or programmatic access method for Gemini Notebook; all evidence describes UI-driven features (sources, chat, audio/video overviews, sharing) and community commentary about workflow limitations, not API access.

   ⚑ flag

4. ai-native userBuild against official SDKs

   weight 2 · round drawn
   Sakana Marlin—none0/10

   No evidence of any official SDK, API, or developer library for Marlin; probes for llms.txt and OpenAPI specs both returned 404, and all docs describe an end-user research product with no mention of programmatic/SDK access.

   • [probe] “PROBE llms.txt: HTTP 404 at https://sakana.ai/llms.txt”
   • [probe] “PROBE openapi: all candidate paths 404 (https://sakana.ai/openapi.json, https://sakana.ai/swagger.json, https://sakana.ai/api/openapi.json, …”

   proof ↗⚑ flag

   Gemini Notebook (NotebookLM)—none0/10

   No evidence of an official SDK, API, or developer library for building against Gemini Notebook; all evidence describes end-user features (uploading sources, chat, audio/video overviews, notes) rather than programmatic/developer access.

   ⚑ flag

5. ai-native userSubscribe to events via webhooks

   weight 2 · round drawn
   Sakana Marlin—none0/10

   No evidence of a webhook or event subscription mechanism; probes for API/OpenAPI specs returned 404s and docs focus only on research report generation. Missing for 10: any webhook documentation, event subscription API, or callback mechanism.

   • [probe] “PROBE llms.txt: HTTP 404 at https://sakana.ai/llms.txt”
   • [probe] “PROBE openapi: all candidate paths 404 (https://sakana.ai/openapi.json, https://sakana.ai/swagger.json, https://sakana.ai/api/openapi.json, …”

   proof ↗⚑ flag

   Gemini Notebook (NotebookLM)—none0/10

   No evidence of any webhook subscription, API event, or outbound notification capability in Gemini Notebook's docs or community feedback; the product is a research/notebook tool with no mentioned event-driven integration surface.

   ⚑ flag

Agentic features

1. ai-native userGet AI-generated insights and suggestions from my data inside the product

   weight 2 · round drawn
   Sakana Marlin✓fullclaimed8/10

   Marlin autonomously researches user-provided topics, mapping causal relationships, comparing hypotheses, and generating structured strategic insights and reports with citations, going beyond simple summarization. Missing for 10: independent/hands-on third-party validation beyond vendor-curated testimonials, and no visibility into underlying data/insight quality benchmarks.

   • [claimed-docs] “Sakana Marlin maps the causal relationships at work in complex business environments and organizes them into structured strategic options.”
   • [claimed-docs] “It does more than summarize. Sakana Marlin maps the causal relationships at work in complex business environments and organizes them into st…”
   • [claimed-docs] “Rather than merely aggregating information, it compares and evaluates multiple hypotheses to provide deep insights.”
   • [claimed-docs] “From ideation and data gathering to resolving contradictions and structuring the report, the AI autonomously executes the entire research pr…”
   • [claimed-docs] “The research was of extremely high quality, grounded strictly in primary sources, resulting in a highly convincing and reliable final report…”
   • [claimed-docs] “Give it a research topic, and Marlin works autonomously for up to roughly eight hours, crafting a detailed strategy report up to a hundred p…”

   proof ↗⚑ flag

   Gemini Notebook (NotebookLM)✓fullcommunity8/10

   NotebookLM generates summaries, insights, mind maps, briefing documents, AI-suggested report types, and answers with grounded citations directly from uploaded sources, and community feedback confirms real-world use for research insight extraction (e.g., audio overviews, code walkthroughs). Missing for 10: independent benchmarking of insight quality/accuracy and evidence of more proactive/agentic suggestion behavior beyond user-initiated Q&A.

   • [claimed-docs] “Chat with your notebook to get grounded information based on your sources with clear in-line citations for accuracy, transparency, and trust…”
   • [claimed-docs] “Find an auto-generated summary of the entire source in the Source Guide.”
   • [claimed-docs] “Create your own, or choose from FAQ, study guide, briefing document, or an AI suggested report type.”
   • [claimed-docs] “Get insights in the moment by asking questions about your sources.”
   • [claimed-docs] “Mind Maps in Gemini Notebook visually summarize your uploaded sources, showing main topics and related ideas as a branching diagram.”
   • [community] “NotebookLM is genuinely useful for structured research workflows. My concern is not the name. It's whether the "notebook" metaphor survives …”
   • [community] “I used it to listen to a high level walkthrough of interesting and entertaining code while driving to work... Hearing it talk about Ken Thom…”

   proof ↗⚑ flag

2. ai-native userSet up automations that run autonomously in the background

   weight 2 · round to Sakana Marlin
   Sakana Marlin~partialclaimed6/10

   Marlin's docs clearly describe a single research task running autonomously for up to ~8 hours without further human input once a topic is set (docs-9, docs-10, docs-16), which matches the 'runs in background autonomously' idea. However, this is a one-shot session, not a recurring/scheduled automation you configure and forget — there's no evidence of triggers, schedules, or multi-run automation management typical of 'set up automations.' Missing for 10: scheduled/recurring automation setup, background job management UI, independent hands-on corroboration of unattended runtime.

   • [claimed-docs] “Give it a research topic, and Marlin works autonomously for up to roughly eight hours, crafting a detailed strategy report up to a hundred p…”
   • [claimed-docs] “Sakana Marlin sharpens the direction of the investigation through a brief exchange with the user. Once the course is set, it works without f…”
   • [claimed-docs] “From ideation and data gathering to resolving contradictions and structuring the report, the AI autonomously executes the entire research pr…”
   • [claimed-docs] “It is designed for use cases where you want to delegate everything up to "issue framing (agenda setting)" prior to decision-making to the AI…”

   proof ↗⚑ flag

   Gemini Notebook (NotebookLM)—none0/10

   No evidence of scheduling, triggers, or autonomous background automations in NotebookLM's documented features; all capabilities (chat, source summarization, audio/video overviews, notes) are user-initiated within a session.

   ⚑ flag

3. ai-native userDelegate tasks to a built-in AI assistant inside the product

   weight 3 · round to Sakana Marlin
   Sakana Marlin✓fullclaimed8/10

   Marlin is explicitly designed as a built-in AI agent that users delegate entire research/strategy tasks to, working autonomously for hours with minimal human input beyond initial framing (docs-1, docs-9, docs-10, docs-16, docs-21). This directly matches the story of delegating tasks to a built-in assistant within the product. Missing for 10: independent/hands-on verification beyond vendor testimonials, and detail on interactive control while a task is delegated.

   • [claimed-docs] “Once you set a research topic, it works without further human input: repeatedly forming hypotheses, gathering information, navigating the we…”
   • [claimed-docs] “Give it a research topic, and Marlin works autonomously for up to roughly eight hours, crafting a detailed strategy report up to a hundred p…”
   • [claimed-docs] “Sakana Marlin sharpens the direction of the investigation through a brief exchange with the user. Once the course is set, it works without f…”
   • [claimed-docs] “From ideation and data gathering to resolving contradictions and structuring the report, the AI autonomously executes the entire research pr…”
   • [claimed-docs] “It is designed for use cases where you want to delegate everything up to "issue framing (agenda setting)" prior to decision-making to the AI…”

   proof ↗⚑ flag

   Gemini Notebook (NotebookLM)~partialcommunity4/10

   NotebookLM lets users chat with sources and generate transformed outputs (summaries, audio/video overviews, study guides, slide decks), which is a form of task delegation, but there is no evidence of an agentic assistant that can take autonomous multi-step actions, use tools, browse/search independently, or execute tasks beyond source-grounded Q&A and content transformation; community feedback even notes it can't search for docs on its own (notebooklm-comm-4). missing for 10: autonomous multi-step task execution, tool/agent orchestration, independent web search/retrieval, evidence of true delegation beyond content generation.

   • [claimed-docs] “Chat with your notebook to get grounded information based on your sources with clear in-line citations for accuracy, transparency, and trust…”
   • [claimed-docs] “Transform your sources into approachable formats such as study guides, briefings, audio overviews, mind maps, and more.”
   • [claimed-docs] “Audio Overviews are deep-dive discussions between AI hosts that provide in-depth summaries of the key topics in your uploaded sources.”
   • [claimed-docs] “Transform your sources into a polished presentation by generating a Slide Deck, perfect for sharing or reviewing key concepts.”
   • [community] “My problem with NotebookLM was that you had to give it URLs or docs, you couldn't just have it search for docs and then incorporate them int…”

   proof ↗⚑ flag

4. ai-native userOperate the product with natural-language commands

   weight 2 · round to Gemini Notebook (NotebookLM)
   Sakana Marlin~partialclaimed5/10

   Marlin is initiated by giving it a research topic and a brief natural-language exchange to set direction (docs-10), suggesting natural-language input drives its operation, but there is no evidence of a broader natural-language command interface (e.g., chat-style control, follow-up instructions, or command syntax) beyond initial topic-setting. Missing for 10: documentation of ongoing NL command/control during execution, examples of varied NL commands, independent/hands-on confirmation of NL interaction quality.

   • [claimed-docs] “Sakana Marlin sharpens the direction of the investigation through a brief exchange with the user. Once the course is set, it works without f…”
   • [claimed-docs] “Give it a research topic, and Marlin works autonomously for up to roughly eight hours, crafting a detailed strategy report up to a hundred p…”
   • [claimed-docs] “It is designed for use cases where you want to delegate everything up to "issue framing (agenda setting)" prior to decision-making to the AI…”

   proof ↗⚑ flag

   Gemini Notebook (NotebookLM)~partialcommunity6/10

   NotebookLM lets users interact via natural-language chat to query sources, request custom conversational styles/roles, and ask questions (notebooklm-docs-2, notebooklm-docs-7, notebooklm-docs-28), which is a genuine natural-language interface. However, many core actions (uploading sources, converting notes, generating mind maps/slide decks/audio overviews) are triggered via UI menus/buttons rather than free-form commands, and community feedback notes limits like being unable to have it search for and incorporate new sources on its own (notebooklm-comm-4). Missing for 10: evidence of natural-language commands driving broader agentic actions (e.g., source discovery, format generation, sharing) rather than just chat Q&A and style prompts, and independent confirmation of command robustness.

   • [claimed-docs] “Chat with your notebook to get grounded information based on your sources with clear in-line citations for accuracy, transparency, and trust…”
   • [claimed-docs] “Custom: Best for choosing a specific style ("Respond like a PhD student") or suggesting a role ("Pretend to be a role-playing game host.").”
   • [claimed-docs] “Get insights in the moment by asking questions about your sources.”
   • [community] “My problem with NotebookLM was that you had to give it URLs or docs, you couldn't just have it search for docs and then incorporate them int…”

   proof ↗⚑ flag

Api quality

1. ai-native userDownload a machine-readable API spec (OpenAPI or equivalent)

   weight 2 · round drawn
   Sakana Marlin—none0/10

   Probes for OpenAPI/swagger spec files and llms.txt all returned 404, and no documentation mentions a machine-readable API spec. Missing for 10: any published OpenAPI/Swagger file, API reference docs, or equivalent machine-readable spec.

   • [probe] “PROBE llms.txt: HTTP 404 at https://sakana.ai/llms.txt”
   • [probe] “PROBE openapi: all candidate paths 404 (https://sakana.ai/openapi.json, https://sakana.ai/swagger.json, https://sakana.ai/api/openapi.json, …”

   proof ↗⚑ flag

   Gemini Notebook (NotebookLM)—none0/10

   No evidence anywhere in the pack of an OpenAPI spec, API documentation, or any machine-readable API for Gemini Notebook; all evidence covers UI features (sources, notes, audio/video overviews, sharing). missing for 10: any mention of a public API, OpenAPI/Swagger spec, or SDK.

   ⚑ flag

2. ai-native userRely on versioned APIs with a documented deprecation policy

   weight 2 · round drawn
   Sakana Marlin—none0/10

   No evidence of any public API, versioning scheme, or deprecation policy; probes for OpenAPI/llms.txt endpoints returned 404s, and all docs describe the research product itself, not a developer API. Missing for 10: any API documentation, versioning scheme, deprecation policy.

   • [probe] “PROBE llms.txt: HTTP 404 at https://sakana.ai/llms.txt”
   • [probe] “PROBE openapi: all candidate paths 404 (https://sakana.ai/openapi.json, https://sakana.ai/swagger.json, https://sakana.ai/api/openapi.json, …”

   proof ↗⚑ flag

   Gemini Notebook (NotebookLM)—none0/10

   No evidence of a versioned public API for NotebookLM/Gemini Notebook, let alone a documented deprecation policy; all evidence covers end-user features (uploads, summaries, audio/video overviews) not developer API access. Missing for 10: any mention of an API, versioning scheme, or deprecation/support-lifecycle policy.

   ⚑ flag

1. ai-native userPerform bulk operations across many items at once

   weight 2 · round drawn
   Sakana Marlin—none0/10

   All evidence describes Marlin running a single autonomous research task per topic (one report, one investigation) rather than processing many items/topics in bulk; no mention of batch submission, multi-item queues, or parallel bulk execution.

   ⚑ flag

   Gemini Notebook (NotebookLM)—none0/10

   The evidence only shows per-source actions (uploading individual sources, converting a single note to a source, generating one summary or audio overview at a time) with no mention of bulk/batch operations across many sources, notebooks, or items simultaneously. There is no documented bulk-select, bulk-delete, bulk-export, or batch-processing capability anywhere in the docs or community feedback.

   • [claimed-docs] “Easily upload PDFs, websites, YouTube videos, audio files, Google Docs, or Google Slides or discover new sources.”
   • [claimed-docs] “To convert a single note to a source: Open the note. Select Convert to source.”
   • [claimed-docs] “Each source can contain up to 500,000 words or up to 200MB for uploaded files.”

   proof ↗⚑ flag

2. ai-native userSchedule recurring jobs or workflows

   weight 2 · round drawn
   Sakana Marlin—none0/10

   Sakana Marlin is a single-run autonomous research/report tool triggered by a user topic; no evidence of scheduling, recurrence, cron-like triggers, or workflow automation for repeated jobs. Probes also show no API/OpenAPI surface that could support scheduled invocation.

   • [claimed-docs] “Give it a research topic, and Marlin works autonomously for up to roughly eight hours, crafting a detailed strategy report up to a hundred p…”
   • [claimed-docs] “Sakana Marlin sharpens the direction of the investigation through a brief exchange with the user. Once the course is set, it works without f…”
   • [probe] “PROBE llms.txt: HTTP 404 at https://sakana.ai/llms.txt”
   • [probe] “PROBE openapi: all candidate paths 404 (https://sakana.ai/openapi.json, https://sakana.ai/swagger.json, https://sakana.ai/api/openapi.json, …”

   proof ↗⚑ flag

   Gemini Notebook (NotebookLM)—none0/10

   No evidence anywhere in the pack of scheduling, recurring jobs, triggers, or automated workflow execution; NotebookLM is a research/notebook tool driven by manual source uploads and on-demand queries.

   ⚑ flag

Sharing

1. analystShare a research session or report with collaborators who can view or build on it

   weight 2 · round to Gemini Notebook (NotebookLM)
   Sakana Marlin—none0/10

   No evidence describes any collaboration or sharing features—no mention of shared workspaces, links, comments, or multi-user access to reports/sessions; the pack only covers autonomous research generation, pricing tiers, and API probes returning 404s.

   • [claimed-docs] “We have made Marlin available as a pay-per-use tier to monthly Pro, Team, and Enterprise-tier plans.”
   • [probe] “PROBE llms.txt: HTTP 404 at https://sakana.ai/llms.txt”
   • [probe] “PROBE openapi: all candidate paths 404 (https://sakana.ai/openapi.json, https://sakana.ai/swagger.json, https://sakana.ai/api/openapi.json, …”

   proof ↗⚑ flag

   Gemini Notebook (NotebookLM)~partialclaimed7/10

   Docs confirm notebooks can be made public via shareable link, that Workspace admins now treat Gemini Notebook as a core service (implying org-level sharing), and that notes/slide decks/reports can be exported to Google Docs/Sheets or PDF so collaborators can view and continue working on them. However, evidence only describes public link sharing and analytics, not granular collaborator invite/edit permissions typical of collaborative docs, and there's no independent/hands-on account of real-world collaboration workflows. Missing for 10: explicit collaborator role/permission controls (view vs edit), independent corroboration of multi-user collaboration in practice.

   • [claimed-docs] “You can make your notebook public and share a link with anyone with a Google account.”
   • [claimed-docs] “Owners of public notebooks with paid subscriptions can view usage analytics in the analytics tab.”
   • [claimed-docs] “You can export your notes to Google Docs or Google Sheets.”
   • [claimed-docs] “For Workspace and Workspace for Education users, Gemini Notebook is now a core service (like Gmail, Google Calendar, and Docs).”
   • [claimed-docs] “Transform your sources into a polished presentation by generating a Slide Deck, perfect for sharing or reviewing key concepts. Present your …”

   proof ↗⚑ flag

Alerts

1. researcherSet up standing searches or alerts that surface new relevant sources as they appear

   weight 1 · round drawn
   Sakana Marlin—none0/10

   Marlin is a one-shot deep-research report generator that runs autonomously for a single session (up to ~8 hours) and produces a report; there is no evidence of standing searches, saved alerts, or recurring monitoring for new sources over time.

   ⚑ flag

   Gemini Notebook (NotebookLM)—none0/10

   No evidence of any standing-search/alert capability that surfaces new relevant sources automatically; sources must be manually uploaded or discovered by the user. Community feedback explicitly confirms this gap: 'you had to give it URLs or docs, you couldn't just have it search for docs and then incorporate them into your research' ([notebooklm-comm-4]).

   • [community] “My problem with NotebookLM was that you had to give it URLs or docs, you couldn't just have it search for docs and then incorporate them int…”
   • [claimed-docs] “Easily upload PDFs, websites, YouTube videos, audio files, Google Docs, or Google Slides or discover new sources.”

   proof ↗⚑ flag

Corpus

1. researcherUpload my own PDFs or corpus and have the agent research over them

   weight 2 · round to Gemini Notebook (NotebookLM)
   Sakana Marlin—none0/10

   Evidence describes Marlin as an autonomous web-research agent that gathers information via web navigation and generates reports, but there is no mention of uploading a user's own PDFs or corpus for the agent to research over. Missing for 10: any document/file upload feature, corpus ingestion, or evidence of researching over user-supplied materials rather than open web sources.

   ⚑ flag

   Gemini Notebook (NotebookLM)✓fullcommunity8/10

   Docs confirm PDF/document upload (up to 500k words/200MB per source), grounded Q&A with inline citations navigable to source location, source guide summaries, and note-taking over uploaded corpora—directly matching the story. Community evidence corroborates real-world use for structured research and studying, though one comment notes it can't proactively search for and pull in new docs beyond what's uploaded. Missing for 10: independent benchmarking of research quality/accuracy across large multi-PDF corpora, and confirmation of robust handling of very large or messy academic corpora at scale.

   • [claimed-docs] “Easily upload PDFs, websites, YouTube videos, audio files, Google Docs, or Google Slides or discover new sources.”
   • [claimed-docs] “Chat with your notebook to get grounded information based on your sources with clear in-line citations for accuracy, transparency, and trust…”
   • [claimed-docs] “Each source can contain up to 500,000 words or up to 200MB for uploaded files.”
   • [claimed-docs] “Find an auto-generated summary of the entire source in the Source Guide.”
   • [claimed-docs] “You can hover over any citation to get the full quoted text right away.”
   • [claimed-docs] “If you select a citation, Gemini Notebook automatically navigates to the location of the quote, so you can easily view it in context.”
   • [claimed-docs] “You can use notes to capture and organize information, identify key insights and interpretations from your sources, and add your own thought…”
   • [community] “NotebookLM is genuinely useful for structured research workflows. My concern is not the name. It's whether the "notebook" metaphor survives …”
   • [community] “My problem with NotebookLM was that you had to give it URLs or docs, you couldn't just have it search for docs and then incorporate them int…”

   proof ↗⚑ flag

Reviews

1. researcherRun a systematic screening and extraction workflow across many papers with consistent criteria

   weight 2 · round to Gemini Notebook (NotebookLM)
   Sakana Marlin—none0/10

   Marlin is positioned as an autonomous business/market strategy research agent producing single deep-dive reports, not as a tool for systematic multi-paper screening/extraction with consistent criteria (a literature-review workflow). No evidence describes handling many papers, applying consistent inclusion/extraction criteria, or batch processing across a corpus — the described unit of work is one topic producing one report.

   • [claimed-docs] “Give it a research topic, and Marlin works autonomously for up to roughly eight hours, crafting a detailed strategy report up to a hundred p…”
   • [claimed-docs] “It is designed for use cases where you want to delegate everything up to "issue framing (agenda setting)" prior to decision-making to the AI…”
   • [claimed-docs] “putting Marlin to work on real tasks such as strategy formulation, market research, risk analysis, and competitive analysis”

   proof ↗⚑ flag

   Gemini Notebook (NotebookLM)~partialcommunity3/10

   NotebookLM supports uploading many sources into one notebook and chatting with grounded, cited answers, which could be used ad hoc to ask the same question across sources, but there is no evidence of a systematic screening/extraction feature (e.g., structured criteria templates, per-paper data tables, PRISMA-style tracking, or bulk batch processing) needed for a repeatable literature review workflow. Community feedback also notes the product cannot auto-search/pull in new sources itself, requiring manual curation, which undercuts a large-scale systematic workflow. Missing for 10: structured extraction templates/tables applied consistently across many papers, evidence of handling large paper sets with defined screening criteria, and confirmation this scales beyond source-count limits.

   • [claimed-docs] “Chat with your notebook to get grounded information based on your sources with clear in-line citations for accuracy, transparency, and trust…”
   • [claimed-docs] “You can hover over any citation to get the full quoted text right away.”
   • [claimed-docs] “Get insights in the moment by asking questions about your sources.”
   • [community] “My problem with NotebookLM was that you had to give it URLs or docs, you couldn't just have it search for docs and then incorporate them int…”

   proof ↗⚑ flag

1. ai-native userDo everything through the API that I can do in the UI

   weight 2 · round drawn
   Sakana Marlin—none0/10

   No evidence of any public API for Sakana Marlin; probes for llms.txt and OpenAPI spec both returned 404, and all documentation describes only UI/credit-based access.

   • [probe] “PROBE llms.txt: HTTP 404 at https://sakana.ai/llms.txt”
   • [probe] “PROBE openapi: all candidate paths 404 (https://sakana.ai/openapi.json, https://sakana.ai/swagger.json, https://sakana.ai/api/openapi.json, …”
   • [claimed-docs] “Add your card. Start right away.”

   proof ↗⚑ flag

   Gemini Notebook (NotebookLM)—none0/10

   No evidence pack items mention any public API, developer SDK, or programmatic access to Gemini Notebook features; all evidence describes UI-only workflows (upload, chat, generate audio/video/slides, notes, sharing). Community comments only discuss UI limitations, not API availability. missing for 10: any mention of a public API, SDK, or programmatic endpoints; documentation of API parity with UI features; developer access/authentication details.

   ⚑ flag

2. ai-native userExport all of my data in open formats and leave

   weight 3 · round to Gemini Notebook (NotebookLM)
   Sakana Marlin—none0/10

   No evidence of data export, open-format download, or account portability features; probes for llms.txt and OpenAPI both returned 404, and docs only describe generated reports/slides, not export of underlying user data.

   • [probe] “PROBE llms.txt: HTTP 404 at https://sakana.ai/llms.txt”
   • [probe] “PROBE openapi: all candidate paths 404 (https://sakana.ai/openapi.json, https://sakana.ai/swagger.json, https://sakana.ai/api/openapi.json, …”

   proof ↗⚑ flag

   Gemini Notebook (NotebookLM)~partialclaimed3/10

   Docs show only piecemeal export options—notes to Google Docs/Sheets, audio overview downloads, and slide decks as PDF—but no evidence of a full data export (sources, chat history, citations, mind maps) or a genuine 'leave the platform' workflow in open, portable formats. Missing for 10: bulk/full-notebook export, open-format guarantees (vs Google Docs/Sheets lock-in), and any documentation or community confirmation of a complete account data export/exit path.

   • [claimed-docs] “You can export your notes to Google Docs or Google Sheets.”
   • [claimed-docs] “Listen to Audio Overviews on the go and download them for offline playback.”
   • [claimed-docs] “Transform your sources into a polished presentation by generating a Slide Deck, perfect for sharing or reviewing key concepts. Present your …”

   proof ↗⚑ flag

Pricing

1. researcherTry the product meaningfully on a free tier or trial

   weight 1 · round to Gemini Notebook (NotebookLM)
   Sakana Marlin—none0/10

   Marlin is explicitly pay-per-use available only to paid Pro/Team/Enterprise plans, requires adding a card to start, and cancelling mid-run still consumes credits — there is no free tier or trial for researchers to test it meaningfully.

   • [claimed-docs] “We have made Marlin available as a pay-per-use tier to monthly Pro, Team, and Enterprise-tier plans.”
   • [claimed-docs] “Add your card. Start right away.”
   • [claimed-docs] “You can cancel at any time during execution, but please note that credits will still be consumed.”
   • [claimed-docs] “Add-on credits available (¥98 / credit)”

   proof ↗⚑ flag

   Gemini Notebook (NotebookLM)~partialcommunity4/10

   Evidence confirms a paid subscription tier exists (analytics for paid users) implying a separate free tier, and community comments describe hands-on use (audio overviews, code walkthroughs) without mention of payment, suggesting meaningful free access. However, no doc explicitly states free-tier quotas, source limits, or trial duration for researchers. Missing for 10: explicit free-tier usage caps/limits documentation, direct confirmation of no-cost access, and independent confirmation of research-scale usability without upgrading.

   • [claimed-docs] “Owners of public notebooks with paid subscriptions can view usage analytics in the analytics tab.”
   • [community] “I used it to listen to a high level walkthrough of interesting and entertaining code while driving to work... Hearing it talk about Ken Thom…”
   • [community] “Much better name for this well-designed product. I do wish they made it easier to pull in emails. I had to use gyb and mhonarc to retrieve e…”

   proof ↗⚑ flag

2. researcherUnderstand plan pricing and usage limits before committing

   weight 2 · round to Sakana Marlin
   Sakana Marlin~partialclaimed4/10

   Marlin's docs mention it's offered as a pay-per-use add-on to Pro/Team/Enterprise plans, credits are consumed even if cancelled mid-run, and additional credits cost ¥98 each, giving a researcher some pricing signal. However, there's no concrete breakdown of how many credits a typical run consumes, no explicit usage caps, and no comparison table of plan tiers — missing for 10: full plan pricing table, credit-consumption-per-task estimates, explicit usage limits, independent/hands-on cost verification.

   • [claimed-docs] “You can cancel at any time during execution, but please note that credits will still be consumed.”
   • [claimed-docs] “We have made Marlin available as a pay-per-use tier to monthly Pro, Team, and Enterprise-tier plans.”
   • [claimed-docs] “Add-on credits available (¥98 / credit)”
   • [claimed-docs] “Add your card. Start right away.”

   proof ↗⚑ flag

   Gemini Notebook (NotebookLM)~partialclaimed3/10

   Docs mention some usage limits (500,000 words/200MB per source) and note that usage analytics require a paid subscription and Workspace inclusion, but there is no comprehensive pricing page, tier comparison, or explicit usage caps (e.g., number of notebooks, audio overviews per day) that a researcher would need to fully evaluate cost before committing. Missing for 10: full pricing tier breakdown, free vs paid feature/usage limits, and independent confirmation of these limits.

   • [claimed-docs] “Each source can contain up to 500,000 words or up to 200MB for uploaded files.”
   • [claimed-docs] “Owners of public notebooks with paid subscriptions can view usage analytics in the analytics tab.”
   • [claimed-docs] “For Workspace and Workspace for Education users, Gemini Notebook is now a core service (like Gmail, Google Calendar, and Docs).”

   proof ↗⚑ flag

1. ai-native userChoose where my data is stored (region/residency)

   weight 2 · round drawn
   Sakana Marlin—none0/10

   No evidence in the pack mentions data residency, region selection, or storage location controls for Sakana Marlin; the evidence covers only research capabilities and pricing.

   ⚑ flag

   Gemini Notebook (NotebookLM)—none0/10

   No evidence in the pack mentions data residency, region selection, or storage location controls for Gemini Notebook; the docs cover features like sources, audio/video overviews, and sharing but say nothing about where data is stored or user control over it.

   ⚑ flag

2. ai-native userPrevent my data from being used to train AI models

   weight 3 · round drawn
   Sakana Marlin—none0/10

   No evidence in the pack addresses data-training opt-out, privacy controls, or data usage policy for Sakana Marlin; missing for 10: any mention of training opt-out settings, privacy policy, or data retention controls.

   ⚑ flag

   Gemini Notebook (NotebookLM)—none0/10

   No evidence in the pack addresses data-training opt-out, privacy policy, or data usage controls for Gemini Notebook; all citations concern features (sources, summaries, audio/video overviews) rather than training-data privacy posture.

   ⚑ flag

3. ai-native userControl data retention and deletion

   weight 2 · round drawn
   Sakana Marlin—none0/10

   No evidence in the pack addresses data retention, deletion controls, or privacy settings for Marlin; the evidence pack only covers research capabilities, pricing, and product marketing.

   ⚑ flag

   Gemini Notebook (NotebookLM)—none0/10

   No evidence in the pack addresses data retention controls, deletion policies, or user ability to manage/export/delete stored data and sources beyond basic note export; the axis clearly applies to an AI-native privacy-conscious user but no documentation or community mention covers retention/deletion controls.

   ⚑ flag

4. ai-native userOpt out of telemetry and usage tracking

   weight 2 · round drawn
   Sakana Marlin—none0/10

   No evidence in the pack addresses telemetry, usage tracking, or any privacy/opt-out controls for Sakana Marlin. Missing for 10: any documentation of telemetry practices, an opt-out mechanism, or privacy policy details.

   ⚑ flag

   Gemini Notebook (NotebookLM)—none0/10

   No evidence pack item mentions telemetry, usage-tracking opt-out, privacy controls, or data-collection settings; only feature docs and general community commentary are present.

   ⚑ flag

Reports

1. researcherExport results to common formats, including documents, spreadsheets, and reference-manager files

   weight 1 · round to Gemini Notebook (NotebookLM)
   Sakana Marlin~partialclaimed4/10

   Marlin auto-generates full reports with references and PowerPoint slides (docs-3, docs-9, docs-20), covering the 'documents' part of the story, but there is no evidence of spreadsheet export or reference-manager file formats (e.g., BibTeX/RIS/EndNote) for citations. missing for 10: spreadsheet export, reference-manager file export (BibTeX/RIS/EndNote), independent corroboration of export formats.

   • [claimed-docs] “Everything from the main body to appendices, references, and presentation slides is automatically generated — at a quality that stands along…”
   • [claimed-docs] “Give it a research topic, and Marlin works autonomously for up to roughly eight hours, crafting a detailed strategy report up to a hundred p…”
   • [claimed-docs] “From a fully referenced report to PowerPoint slides, everything is automatically generated — at a quality that stands alongside professional…”

   proof ↗⚑ flag

   Gemini Notebook (NotebookLM)~partialclaimed5/10

   NotebookLM supports exporting notes to Google Docs or Google Sheets and can generate a Slide Deck exportable as PDF, covering document/spreadsheet-style outputs, but there is no evidence of export to reference-manager formats (e.g., BibTeX, RIS, EndNote) which the story explicitly requires. missing for 10: reference-manager file export, independent corroboration of export fidelity.

   • [claimed-docs] “You can export your notes to Google Docs or Google Sheets.”
   • [claimed-docs] “Transform your sources into a polished presentation by generating a Slide Deck, perfect for sharing or reviewing key concepts. Present your …”

   proof ↗⚑ flag

2. analystGet a structured report with sections, tables, and a summary that I can share with stakeholders

   weight 3 · round to Sakana Marlin
   Sakana Marlin✓fullclaimed7/10

   Docs describe autonomous generation of a full structured report with main body, appendices, references, and presentation slides/executive summary — directly matching sections, tables (implied by structured strategic options), and summary needs for stakeholder sharing. Missing for 10: no independent/hands-on verification of table formatting or actual sample report shown, and no evidence beyond vendor marketing copy.

   • [claimed-docs] “Everything from the main body to appendices, references, and presentation slides is automatically generated — at a quality that stands along…”
   • [claimed-docs] “Give it a research topic, and Marlin works autonomously for up to roughly eight hours, crafting a detailed strategy report up to a hundred p…”
   • [claimed-docs] “From a fully referenced report to PowerPoint slides, everything is automatically generated — at a quality that stands alongside professional…”
   • [claimed-docs] “Sakana Marlin maps the causal relationships at work in complex business environments and organizes them into structured strategic options.”

   proof ↗⚑ flag

   Gemini Notebook (NotebookLM)~partialclaimed6/10

   NotebookLM supports Reports with FAQ/study guide/briefing document/custom formats and auto-generated source summaries, plus export of notes to Google Docs/Sheets and Slide Deck generation for sharing, which together cover structured, shareable outputs. However, evidence doesn't confirm native table generation within reports or a single unified 'report' combining sections+tables+summary in one export artifact. missing for 10: explicit table support in generated reports, a single cohesive exportable report format (vs separate docs/sheets/slides), independent/hands-on validation of report quality for stakeholder sharing.

   • [claimed-docs] “Create your own, or choose from FAQ, study guide, briefing document, or an AI suggested report type.”
   • [claimed-docs] “Find an auto-generated summary of the entire source in the Source Guide.”
   • [claimed-docs] “You can export your notes to Google Docs or Google Sheets.”
   • [claimed-docs] “Transform your sources into a polished presentation by generating a Slide Deck, perfect for sharing or reviewing key concepts.”
   • [claimed-docs] “Transform your sources into a polished presentation by generating a Slide Deck, perfect for sharing or reviewing key concepts. Present your …”

   proof ↗⚑ flag

Agent runs

1. researcherPose a research question and get an autonomous multi-step investigation, not just a single-pass summary

   weight 3 · round to Sakana Marlin
   Sakana Marlin✓fullclaimed8/10

   Vendor docs describe autonomous multi-step research: forming hypotheses, gathering info, resolving contradictions, running for hours across thousands of cycles without further human input, producing a full structured report - directly matching the story. This is corroborated by beta-tester quotes praising depth beyond chat-based research tools, though all evidence is vendor-published/testimonial rather than independent hands-on verification. missing for 10: independent third-party evaluation or benchmark of the autonomous multi-step process, technical detail on how contradictions/hypotheses are actually verified

   • [claimed-docs] “Once you set a research topic, it works without further human input: repeatedly forming hypotheses, gathering information, navigating the we…”
   • [claimed-docs] “Give it a research topic, and Marlin works autonomously for up to roughly eight hours, crafting a detailed strategy report up to a hundred p…”
   • [claimed-docs] “Sakana Marlin sharpens the direction of the investigation through a brief exchange with the user. Once the course is set, it works without f…”
   • [claimed-docs] “From ideation and data gathering to resolving contradictions and structuring the report, the AI autonomously executes the entire research pr…”
   • [claimed-docs] “Many told us that Marlin was more practical at digging deeply into information than the chat-based research tools they had used before”
   • [claimed-docs] “Rather than merely aggregating information, it compares and evaluates multiple hypotheses to provide deep insights.”

   proof ↗⚑ flag

   Gemini Notebook (NotebookLM)—none0/10

   All evidence describes NotebookLM as a source-grounded chat/summarization tool: users upload sources, ask questions, and get single-pass answers, summaries, or generated formats (audio overviews, mind maps, slide decks) — there is no evidence of an autonomous multi-step investigation loop that iteratively searches, plans, or expands beyond the provided sources. A community report explicitly notes it cannot autonomously search for and incorporate new documents into research, reinforcing that it is single-pass/source-bound rather than agentic.

   • [claimed-docs] “Chat with your notebook to get grounded information based on your sources with clear in-line citations for accuracy, transparency, and trust…”
   • [claimed-docs] “Find an auto-generated summary of the entire source in the Source Guide.”
   • [claimed-docs] “Create your own, or choose from FAQ, study guide, briefing document, or an AI suggested report type.”
   • [community] “My problem with NotebookLM was that you had to give it URLs or docs, you couldn't just have it search for docs and then incorporate them int…”

   proof ↗⚑ flag

2. analystStart a long research job that keeps working unattended and notifies me when the result is ready

   weight 2 · round to Sakana Marlin
   Sakana Marlin~partialclaimed6/10

   Vendor docs clearly describe long unattended autonomous research runs (~8 hours) producing full reports, which supports the core of the story, but there is no mention of a notification mechanism when results are ready and no independent/hands-on corroboration beyond marketing copy. missing for 10: evidence of a completion notification/alert feature, independent verification of unattended runtime and reliability, and API/technical docs confirming job-control (start/monitor/cancel) beyond the marketing page.

   • [claimed-docs] “Give it a research topic, and Marlin works autonomously for up to roughly eight hours, crafting a detailed strategy report up to a hundred p…”
   • [claimed-docs] “Sakana Marlin sharpens the direction of the investigation through a brief exchange with the user. Once the course is set, it works without f…”
   • [claimed-docs] “Once you set a research topic, it works without further human input: repeatedly forming hypotheses, gathering information, navigating the we…”
   • [claimed-docs] “You can cancel at any time during execution, but please note that credits will still be consumed.”

   proof ↗⚑ flag

   Gemini Notebook (NotebookLM)—none0/10

   No evidence in the pack describes an asynchronous/background 'long research job' mode with completion notifications; all cited capabilities (chat, summaries, audio/video overviews, flashcards, etc.) are generated within an active session, not unattended long-running jobs with alerts. Missing for 10: any documented async job queue, background processing status, or notification/alert mechanism when a task completes.

   ⚑ flag

3. researcherSteer the depth, effort, and scope of a research run before or while it executes

   weight 1 · round drawn
   Sakana Marlin~partialclaimed4/10

   Marlin only allows a brief initial exchange to set direction before running fully autonomously for up to ~8 hours with no mid-run steering, and there's no documented control over depth/effort/scope parameters (e.g., report length, time budget, source breadth) beyond the initial topic framing; cancellation is possible but not adjustment. missing for 10: mid-execution steering controls, explicit depth/effort/scope parameters or settings, independent evidence of pre-run configurability beyond a 'brief exchange'.

   • [claimed-docs] “Sakana Marlin sharpens the direction of the investigation through a brief exchange with the user. Once the course is set, it works without f…”
   • [claimed-docs] “Give it a research topic, and Marlin works autonomously for up to roughly eight hours, crafting a detailed strategy report up to a hundred p…”
   • [claimed-docs] “You can cancel at any time during execution, but please note that credits will still be consumed.”
   • [claimed-docs] “It is designed for use cases where you want to delegate everything up to "issue framing (agenda setting)" prior to decision-making to the AI…”

   proof ↗⚑ flag

   Gemini Notebook (NotebookLM)~partialclaimed4/10

   NotebookLM lets users pick conversational style/persona (Default, Custom "PhD student" etc.) and choose report/output formats (FAQ, briefing, study guide), which offers some steering of scope and tone, but there is no evidence of controlling depth/effort (e.g., reasoning length, source breadth, iteration count) or adjusting a run mid-execution — outputs like Audio/Video Overviews are generated as fixed artifacts rather than steerable in-progress runs. missing for 10: mid-run steering controls, explicit depth/effort sliders or parameters, ability to expand/narrow scope during execution, independent evidence of these controls working as intended.

   • [claimed-docs] “Custom: Best for choosing a specific style ("Respond like a PhD student") or suggesting a role ("Pretend to be a role-playing game host.").”
   • [claimed-docs] “Select a conversational style. Default: Best for general purpose research and brainstorming tasks.”
   • [claimed-docs] “Create your own, or choose from FAQ, study guide, briefing document, or an AI suggested report type.”
   • [claimed-docs] “Choose a format. Deep Dive (default) for an in-depth conversation. Two hosts unpack and connect topics from your sources in a lively convers…”

   proof ↗⚑ flag

Citations

1. researcherSee citations for every substantive claim so I can verify it against the underlying source

   weight 3 · round to Gemini Notebook (NotebookLM)
   Sakana Marlin~partialclaimed5/10

   Vendor docs claim Marlin generates 'fully referenced reports' grounded in primary sources with appendices and references, and a testimonial praises its higher-quality citations to primary vs secondary sources, suggesting citation support exists. However, there is no independent verification, no example of inline citation format, and no detail on how claims map to sources for auditability. Missing for 10: independent hands-on verification of citation accuracy, example output showing citation linking, and confirmation citations are traceable/clickable to primary sources.

   • [claimed-docs] “The research was of extremely high quality, grounded strictly in primary sources, resulting in a highly convincing and reliable final report…”
   • [claimed-docs] “他の生成AIと比較して、引用される情報の量と質が高く、二次情報ではなく一次情報を参照できている点に優位性を感じました。”
   • [claimed-docs] “From a fully referenced report to PowerPoint slides, everything is automatically generated — at a quality that stands alongside professional…”
   • [claimed-docs] “Everything from the main body to appendices, references, and presentation slides is automatically generated — at a quality that stands along…”

   proof ↗⚑ flag

   Gemini Notebook (NotebookLM)✓fullclaimed9/10

   Docs confirm grounded chat responses come with clear in-line citations, hovering shows the exact quoted text, and clicking a citation navigates directly to the source location for verification — directly matching the story's requirement to verify claims against underlying sources. Community evidence corroborates general usefulness but doesn't specifically contest citation accuracy. Missing for 10: independent hands-on verification of citation accuracy/completeness across diverse source types.

   • [claimed-docs] “Chat with your notebook to get grounded information based on your sources with clear in-line citations for accuracy, transparency, and trust…”
   • [claimed-docs] “You can hover over any citation to get the full quoted text right away.”
   • [claimed-docs] “If you select a citation, Gemini Notebook automatically navigates to the location of the quote, so you can easily view it in context.”

   proof ↗⚑ flag

Corpus

1. researcherSearch scholarly literature and primary sources, not just the open web

   weight 2 · round to Sakana Marlin
   Sakana Marlin~partialclaimed6/10

   Vendor docs claim research is 'grounded strictly in primary sources' and a testimonial notes it references primary rather than secondary information compared to other generative AI tools, but there is no evidence of dedicated scholarly database access (e.g., PubMed, arXiv, JSTOR integration) or citation verification—only general web navigation is described. missing for 10: evidence of scholarly/academic database integration, independent verification of primary-source claim, and details on how it distinguishes scholarly vs open-web sources.

   • [claimed-docs] “The research was of extremely high quality, grounded strictly in primary sources, resulting in a highly convincing and reliable final report…”
   • [claimed-docs] “他の生成AIと比較して、引用される情報の量と質が高く、二次情報ではなく一次情報を参照できている点に優位性を感じました。”
   • [claimed-docs] “Once you set a research topic, it works without further human input: repeatedly forming hypotheses, gathering information, navigating the we…”
   • [claimed-docs] “From ideation and data gathering to resolving contradictions and structuring the report, the AI autonomously executes the entire research pr…”

   proof ↗⚑ flag

   Gemini Notebook (NotebookLM)!disputedcontradicted3/10

   Docs vaguely mention users can 'discover new sources' (notebooklm-docs-1), but there is no evidence of integration with scholarly databases (e.g., PubMed, JSTOR, Google Scholar) for searching primary literature, and a hands-on community report explicitly states you cannot have NotebookLM search for and incorporate sources itself—you must supply URLs or docs yourself (notebooklm-comm-4). This directly contradicts the implied 'discover' capability, making it a disputed claim rather than a confirmed feature. missing for 10: evidence of any scholarly/primary-source search integration, first-party documentation describing how 'discover sources' actually searches literature, and independent confirmation that discovery works as claimed.

   • [claimed-docs] “Easily upload PDFs, websites, YouTube videos, audio files, Google Docs, or Google Slides or discover new sources.”
   • [community] “My problem with NotebookLM was that you had to give it URLs or docs, you couldn't just have it search for docs and then incorporate them int…”

   proof ↗⚑ flag

Synthesis

1. analystSee where sources agree and disagree instead of a single unqualified answer

   weight 2 · round to Sakana Marlin
   Sakana Marlin~partialclaimed4/10

   Marketing copy claims Marlin 'resolves contradictions' and 'compares and evaluates multiple hypotheses' rather than merely aggregating, implying some handling of conflicting sources, but there is no evidence of a UI feature or report section that explicitly surfaces where sources agree/disagree to the analyst. Missing for 10: concrete example of a report showing conflicting source viewpoints, screenshot/description of how disagreement is presented, independent corroboration beyond vendor marketing.

   • [claimed-docs] “Once you set a research topic, it works without further human input: repeatedly forming hypotheses, gathering information, navigating the we…”
   • [claimed-docs] “From ideation and data gathering to resolving contradictions and structuring the report, the AI autonomously executes the entire research pr…”
   • [claimed-docs] “Rather than merely aggregating information, it compares and evaluates multiple hypotheses to provide deep insights.”

   proof ↗⚑ flag

   Gemini Notebook (NotebookLM)—none0/10

   The evidence describes grounded chat with citations, summaries, mind maps, and study aids, but nothing about explicitly surfacing agreement or contradiction across multiple sources. This is a fair capability to expect from a multi-source research tool, but no docs or community evidence mention comparative or conflict-detection analysis between sources.

   ⚑ flag