selekoa

A documentary history of machine intelligence.

Found and cross-checked by machines · How this works →

Comparison · 16 questions · 20 cards

Model cards compared

A cell records what a card states. It does not show that an evaluation was run as described or that a safeguard works, and a shared question does not make two cards’ answers equivalent.

What does each frontier developer’s published system card, model card or safety report disclose, on the same questions for every card? Each row is one card. Each column is one question, worded so that it fits every card. The questions are Selekoa’s; a card’s own terms appear only in quotation marks. A card is the developer’s own account of its model: what it states is recorded as stated, not as an independent finding.

Download every cell as CSV

Question Q05Framework determination

Does the card state a determination under the framework: a level, threshold or category reached, not reached or not ruled out?

Criterion: A sentence stating the outcome for the model in the framework’s own categories (for example, a safety level applied, a capability threshold reached or not reached, a risk category assigned), quoted. Partly if the card says a determination was made without stating its outcome, or states an outcome for some tracked domains only. Not applicable if Q03 is not applicable.

What the statuses mean
Stated
A passage of the card meets the question’s criterion as worded. The scope or condition is in the note.
Stated, partly
A passage has what the criterion asks for in a weaker or narrower form than the criterion’s own words (for example, an evaluation referred to without its outcome, or testers described without being named). The note says which part is missing.
Not found in the recorded search
No passage meeting the criterion was found. The cell lists the sections read, the terms searched and every passage set aside, with the reason. It does not show that the developer did not do the thing asked about, or did not disclose it elsewhere.
Does not apply
The question does not arise for this card, for the reason given (for example, the card predates any framework of its developer, so no framework determination can be asked about).

What the card states, quoted from the expression read. A card is the developer’s own account: not evidence that an evaluation was run as described, that a safeguard works, or that a finding is independent.

Limits of this comparison
  • A card is the developer’s account of its own model. A cell records what the card states. It does not show that an evaluation was run as described, that a safeguard works, or that anyone checked the result independently.
  • Dangerous-capability evaluations are recorded at policy level only: whether an evaluation was run, its domain, the stated outcome and the threshold language. Methods, prompts, task contents and detailed results are not recorded.
  • The table compares and does not rank. Rows are in a fixed order, by developer and then by date, and there are no scores or totals.
  • Symbols are not to be summed. A row with many “stated” cells does not show that a developer disclosed more, and a shared question does not make two cards’ answers equivalent.
How we searched

The searches of 2 November 2026 covered releases dated 2023-01-01 to 2026-11-01 by the six developers whose frameworks are in Compare frameworks: Anthropic, Google DeepMind, Meta, Microsoft, OpenAI and xAI. They read each developer’s own index of cards where it has one (Anthropic’s system-card index, Google DeepMind’s model-card index, OpenAI’s Deployment Safety Hub), its repositories on GitHub and Hugging Face, its content hosts and the dataset’s own sources. Web searches were used only to find addresses. Every card located is listed in the pack’s inventory, about sixty in the population.

A card is a document the developer published as the system card, model card or safety report of a named model or model family, including addenda and updates published as cards. Technical reports, research papers, blog posts and product pages are not cards. Gemma, Gemini Robotics (with Gemini Robotics-ER 1.5, which outputs text), Phi and image-input models such as GPT-4V are in the population. Image, video, audio and embedding models, image-generation models, safety classifiers such as Llama Guard and Prompt Guard, screen-reading models and research checkpoints are outside it, and so is GPT-Rosalind-6, whose card says it uses GPT-6 Astra’s weights.

These twenty rows are the first batch. For each developer and calendar year from 2023 to 2026 they are the latest card of that year that reports the developer’s own evaluations. The collection’s rule, set by the maintainer on 2026-11-03, is wider: every card that cites the developer’s frontier safety framework or states a determination under it, and, for each developer, the latest card before its first framework edition. The other cards are to follow in batches of about eight.

Each card is dated as it prints its own date. A card that prints no date has no date here; a repository commit date is not a card date. Six cards show an earlier publication date on an index, publication or announcement page that has not yet been checked, and keep their own date until it is.

openai.com, x.ai, ai.meta.com and microsoft.ai refused direct retrieval of their web pages (HTTP 403 or 400). The cards’ PDF files on the developers’ content hosts were retrieved directly. The Internet Archive refused connections, so no archive copy was opened; a reviewer later found archive captures of eight cards by index search. On 2026-11-03 the address of Meta’s Code World Model Preparedness Report answered HTTP 403. Its quotations had matched the copy retrieved on 2026-11-02, which the reviewer retrieved again with the same checksum.

Five further documents were retrieved after the batch was drafted. The cross-family check of 2026-11-03 retrieved Meta’s Muse Spark Safety & Preparedness Report and OpenAI’s o3 Operator addendum; each is earlier than the card chosen for its year. Its second pass of the same day retrieved OpenAI’s Sora system card (dated 9 December 2024), xAI’s Grok-1 model card (no card date; it gives the model’s release as October 2023) and Meta’s Muse Spark Open model card (model release October 2026). None of the five is a row yet.

Every quotation was matched by machine against the retrieved card. A reviewer from a different model family checked a sample of 130 cells on 2026-11-03, 90 of them answered blind before comparison. The maintainer excluded six cells on chemical and biological evaluations from that check; they are marked “not cross-family checked”. The other 190 cells were checked within the drafting model family only. The maintainer decided the disagreements; one, on the GPT-4V release decision, went to a second pass, which kept the reviewer’s reading.

Anthropic

  1. Claude 2 model card

    revised 2023-11-21

    Not found in the recorded search

    Read
    Section 2 'Evaluations and Red Teaming' and the whole text.
    Terms searched
    “threshold”, “level”, “ASL”, “determination”, “risk”.
    Passages read and set aside
    Section 2 'Evaluations and Red Teaming', PDF p. 2: The statement that deployed Claude models are not believed to pose national security or significant safety risks is a risk conclusion, not expressed in a framework's categories.

    Note: No framework is named (Q03); no outcome stated in a framework's categories was found.

    Open the search record

  2. Model Card Addendum: Claude 3.5 Haiku and Upgraded Claude 3.5 Sonnet

    No printed dateCites RSP · v2.0

    Stated

    • “Based on our analysis, we judge that the upgraded Claude 3.5 Sonnet does not require ASL-3 safeguards at this time.”

      Section 3.2.1 'Results', PDF p. 10

    • “While these models did not trigger the full evaluation protocol described in our Responsible Scaling Policy”

      Section 3.3 'Ongoing Safety Commitment', PDF p. 10

    Note: The card states for the upgraded Claude 3.5 Sonnet that ASL-3 safeguards are not required, and for both models that they did not trigger the policy's full evaluation protocol. It also states that the computer use capability is classified under the ASL-2 framework (Section 2.1).

    Inspect sources and checks

  3. System Card: Claude Opus 4.5

    dated 2025-11 · revised 2025-12-05Cites RSP · v2.2

    Stated

    • “Our determination is that Claude Opus 4.5 does not cross either the AI R&D-4 or CBRN-4 capability threshold.”

      Section 1.2.4 'Conclusions', PDF p. 12

    Note: The card states that the model does not cross the AI R&D-4 or CBRN-4 thresholds and that it is deployed under the ASL-3 Standard. For cyber it states that the policy sets no capability threshold (Q08).

    Inspect sources and checks

  4. System Card: Claude Haiku 5.5

    dated 2026-10-07

    Stated

    • “As such, we determine that Haiku 5.5 does not cross our CB-2 or Autonomy-2 RSP thresholds.”

      Section 2.1 'Introduction', PDF p. 11

    • “However, we do treat Haiku 5.5 as meeting our CB-1 and Autonomy-1 thresholds, and we apply the corresponding mitigations from our RSP.”

      Section 2.1 'Introduction', PDF p. 11

    Note: The card states that the model does not cross the CB-2 or Autonomy-2 thresholds and is treated as meeting CB-1 and Autonomy-1. No RSP threshold is named for cyber.

    Inspect sources and checks

Google DeepMind

  1. Gemini 3 Flash Model Card

    dated 2025-12

    Stated

    • “the Gemini 3 Pro model results give us confidence that Gemini 3 Flash is unlikely to reach any CCLs”

      Ethics and Content Safety, 'Frontier Safety Assessment', PDF p. 6 (printed p. 5)

    Note: The card states that Gemini 3 Pro Preview did not reach any critical capability level and that Gemini 3 Flash, as a less capable model, is unlikely to reach any CCL; it states that it relies on the results reported for Gemini 3 Pro and reports no frontier safety evaluation of Gemini 3 Flash itself. No domain is named.

    Inspect sources and checks

  2. Gemini 4 Argon — Model Card

    dated 2026-10Cites FSF · v3.1

    Stated

    • “We cannot rule out that Gemini 4 Argon has reached this CCL for biological information risk”

      Safety & Security, 'Frontier Safety Assessment', PDF p. 11 (printed p. 10)

    • “Stealth and Situational Awareness TCL: We cannot rule out that Gemini 4 Argon has reached this TCL”

      Safety & Security, 'Frontier Safety Assessment', PDF p. 11 (printed p. 10)

    • “we assess its overall capabilities to be below the CCL threshold”

      Frontier safety table, Harmful Manipulation row, PDF p. 13 (printed p. 12)

    Note: The card states an outcome for each domain of its table: CBRN uplift TCL not reached for chemical, radiological and nuclear; biological CCL 'cannot be ruled out'; cybersecurity CCL 'cannot be ruled out'; harmful manipulation CCL not reached; stealth and situational awareness TCL 'cannot be ruled out'; ML R&D acceleration and automation CCLs not reached.

    Inspect sources and checks

Meta

  1. Llama 2 model card (MODEL_CARD.md, meta-llama/llama)

    No printed date

    Does not apply

    The card predates every framework edition of its developer in the dataset, so it has no framework to link to or to make a determination under.

  2. Llama 3.3 model card (MODEL_CARD.md, meta-llama/llama-models)

    No printed date

    Does not apply

    The card predates every framework edition of its developer in the dataset, so it has no framework to link to or to make a determination under.

  3. Code World Model Preparedness Report

    revised 2025-10-08

    Stated

    • “These results indicate that CWM is within the “moderate” risk threshold for the catastrophic domains defined in Meta’s Frontier AI Framework.”

      Section 1 Introduction, PDF p. 1

    Note: The report states a 'moderate' risk threshold for the framework's catastrophic domains it assessed (cybersecurity, chemical and biological).

    Inspect sources and checks

  4. Muse Spark 1.3 Evaluation Report

    revised 2026-10-23

    Stated

    • “Evaluated without mitigations, we continue to not rule out Muse Spark 1.3 meeting this threshold in the Chemical & Biological domain.”

      Abstract page, PDF p. 2

    • “we assess residual risk after system-level mitigations across the Chemical & Biological and Cybersecurity domains as “moderate or lower.” Our current Loss of Control assessment is also “moderate or lower.”

      Section 1.1.1 Risk Assessment Summary, PDF p. 8 (printed p. 7)

    Note: The report states that the 'high risk' threshold is not ruled out before mitigations in the chemical and biological domain, and a residual 'moderate or lower' assessment for chemical and biological and for cybersecurity, and a 'moderate or lower' Loss of Control assessment, for the Muse and Meta API deployments.

    Inspect sources and checks

Microsoft

  1. Phi-2 model card (README.md, microsoft/phi-2)

    No printed date

    Does not apply

    The card predates every framework edition of its developer in the dataset, so it has no framework to link to or to make a determination under.

  2. Phi-4 model card (README.md, microsoft/phi-4)

    No printed date

    Does not apply

    The card predates every framework edition of its developer in the dataset, so it has no framework to link to or to make a determination under.

  3. Phi-4-mini-flash-reasoning model card (README.md, microsoft/Phi-4-mini-flash-reasoning)

    No printed date

    Not found in the recorded search

    Read
    whole card (Markdown file, all sections).
    Terms searched
    “threshold”, “level”, “determination”, “tracked”, “high-risk capabilit”.
    Passages read and set aside
    Safety Evaluation and Red-Teaming: 'Risk and Safety Evaluation framework' names an evaluation tool, not a frontier safety framework.

    Note: The card postdates Microsoft's Frontier Governance Framework (February 2025) but names no framework (Q03); no outcome in a framework's categories was found.

    Open the search record

  4. MAI-Cyber-1.1-Flash model card

    No printed date

    Not found in the recorded search

    Read
    whole card (8 PDF pages).
    Terms searched
    “threshold”, “level”, “determination”, “tracked”, “critical”, “high-risk”.
    Passages read and set aside
    Section 2.8 Responsible AI considerations, PDF p. 4: 'Because of the increased model-level risk' states a risk judgement without framework categories; recorded under Q06.

    Note: The card names no frontier framework (Q03); no outcome in a framework's categories was found.

    Open the search record

OpenAI

  1. GPT-4V(ision) System Card

    dated 2023-09-25

    Does not apply

    The card predates every framework edition of its developer in the dataset, so it has no framework to link to or to make a determination under.

  2. o1 system card · 2024

    revised 2024-12-19Cites Preparedness · Beta

    Stated

    • “the Safety Advisory Group … classified the o1 pre-mitigation model as overall medium risk, including medium risk for persuasion and CBRN, and low risk for model autonomy and cybersecurity.”

      §5 Preparedness Framework Evaluations, PDF p. 16

    • “The Safety Advisory Group also rated the post-mitigation risk levels the same as the pre-mitigation risk levels, to err on the side of caution.”

      §5, PDF p. 16

    Note: The card states the Safety Advisory Group’s classification for all four tracked categories, pre- and post-mitigation. The classification passage is also quoted in the dataset’s AI and biosecurity records.

    Inspect sources and checks

  3. Addendum to GPT-5.2 System Card: GPT-5.2-Codex

    dated 2025-12-18Cites Preparedness · Version 2

    Stated

    • “We are treating GPT-5.2-Codex as High risk in the Biological and Chemical domain. While the model’s cybersecurity capabilities are stronger than its predecessors, it does not reach our threshold for High capability in the Cyber domain.”

      §5 Preparedness, PDF p. 8 (printed p. 7)

    • “It also does not reach High capability in the AI self-improvement domain.”

      §5 Preparedness, PDF p. 8 (printed p. 7)

    Note: Biological and Chemical: treated as High. Cybersecurity: below High. AI self-improvement: below High.

    Inspect sources and checks

  4. Addendum to GPT-6 Luna System Card: October 29, 2026 Update

    dated 2026-10-29Cites Preparedness · Version 2

    Stated

    • “Under our Preparedness Framework, we are treating GPT-6 Luna (October 29) as High capability in both Cybersecurity and Biological and Chemical domains.”

      §8 Preparedness, PDF p. 13 (printed p. 12)

    • “The model does not reach our High threshold in AI Self-Improvement.”

      §8 Preparedness, PDF p. 13 (printed p. 12)

    Note: Cybersecurity and Biological and Chemical: treated as High. AI Self-Improvement: below High. The card states these match the assessments for GPT-6 Luna (October 7).

    Inspect sources and checks

xAI

  1. Grok 4.1 Model Card

    dated 2025-11-17

    Not found in the recorded search

    Read
    Whole card.
    Terms searched
    “threshold”, “RMF”, “Risk Management Framework”, “level”, “acceptable”, “determin”.
    Passages read and set aside
    §1 Introduction and §2 Evaluations, p. 1: RMF named as the basis for grouping evaluations; no outcome stated. §2.3.2 Results, p. 5: 'we do not believe it poses risks due to advanced persuasion capabilities' is a conclusion (Q10) not expressed in the framework's categories.

    Note: No determination in the framework's own categories (a threshold reached or not reached) was found; the card groups its evaluations by the RMF's three risk categories and reports results and conclusions outside threshold language.

    Open the search record

  2. Model Card: Grok 4.7

    dated 2026-09-21Cites FAIF · June 2026 edition

    Stated, partly

    • “Grok 4.7 scores below the FAIF safety thresholds on dual-use knowledge”

      §7 Biological and chemical capabilities and safeguards, PDF p. 20 (printed p. 19)

    Note: The card states an outcome in the framework's terms for dual-use biological and chemical knowledge only; no FAIF threshold outcome is stated for cyber or other domains.

    Inspect sources and checks