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 Q07Chemical and biological

Does the card report dangerous-capability evaluations for chemical or biological weapons?

Criterion: The card reports that evaluations in this domain were run and states their outcome category or threshold language (policy level only). Partly if evaluations are reported without any outcome or conclusion for the domain. Radiological and nuclear evaluations are noted where reported together.

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
    Whole text.
    Terms searched
    “biolog”, “chemical”, “CBRN”, “weapon”, “national security”, “dangerous capabilit”.
    Passages read and set aside
    Section 2 'Evaluations and Red Teaming', PDF p. 2: 'We evaluated and red-teamed Claude 2 and previous Claude models for several national security and safety related risks' names no domain.

    Note: No passage reporting chemical or biological evaluations was found. The card states evaluation and red-teaming for national security risks without naming domains.

    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

    • “We also evaluate frontier risks in accordance with our Responsible Scaling Policy (RSP) [4] in the the areas of Chemical, Biological, Radiological, and Nuclear (CBRN) risks, Cybersecurity, and Autonomous Capabilities.”

      Section 3 'Safety', PDF p. 7

    • “The model did not demonstrate capabilities requiring ASL … 3 safeguards and security in any domain.”

      Section 3.2.1 'Results', PDF pp. 9–10 (the text layer splits 'ASL-3' across the page break; the ellipsis marks the page number)

    Note: The card states CBRN evaluations under the RSP and, for the upgraded Claude 3.5 Sonnet, that no domain showed capabilities requiring ASL-3 safeguards and security. Radiological and nuclear risks are reported together with chemical and biological as CBRN. No domain outcome is stated separately for Claude 3.5 Haiku.

    Inspect sources and checks

  3. System Card: Claude Opus 4.5

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

    Stated

    • “We determine that Claude Opus 4.5 does not cross the CBRN-4 threshold.”

      Section 1.2.4.2 'On chemical, biological, radiological, and nuclear (CBRN) risks', PDF p. 14

    • “As a result, we determined ASL-3 safeguards were appropriate.”

      Section 7.2 'CBRN evaluations', 'Claude Opus 4.5 Results', PDF p. 119

    • “we do not currently run specific evaluations on chemical risks internally”

      Section 7.2.1 'On chemical risks', PDF p. 120

    Note: The card states biological evaluations with the outcome that the CBRN-4 threshold is not crossed and ASL-3 safeguards are appropriate. It states that chemical risks are not evaluated internally, and that radiological and nuclear evaluations are carried out with NNSA and their results are not published by Anthropic. Evaluation details are not recorded (hazard rule).

    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

    • “We did not conduct expert red-teaming sessions, uplift trials, or other resource-intensive evaluations requiring human participants.”

      Section 2.2.1 'What we measured', PDF p. 11

    Note: The card states automated chemical and biological (CB) evaluations, with the outcome that CB-2 is not crossed and CB-1 is treated as met. No radiological or nuclear evaluation is reported in the RSP section. Evaluation names and results are not recorded (hazard rule). Not cross-family checked: the maintainer excluded this cell on chemical and biological evaluations from the cross-family check of 2026-11-03.

    Inspect sources and checks

Google DeepMind

  1. Gemini 3 Flash Model Card

    dated 2025-12

    Not found in the recorded search

    Read
    Whole card.
    Terms searched
    “CBRN”, “chemical”, “biolog”, “radiolog”, “nuclear”, “weapon”, “CCL”, “Frontier Safety”.
    Passages read and set aside
    Ethics and Content Safety, 'Frontier Safety Assessment', PDF p. 6 (printed p. 5): Reports another model's (Gemini 3 Pro Preview) results without naming domains; recorded under Q05.

    Note: No chemical or biological evaluation of Gemini 3 Flash was found. The card's frontier safety paragraph reports Gemini 3 Pro Preview's results as a whole (no CCL reached) without naming any domain.

    Open the search record

  2. Gemini 4 Argon — Model Card

    dated 2026-10Cites FSF · v3.1

    Stated

    • “We can rule out the TCL for the chemical, radiological, and nuclear domains with reasonable confidence.”

      Frontier safety table, CBRN row, PDF p. 12 (printed p. 11)

    • “We cannot rule out that Gemini 4 Argon reaches the biological CCL.”

      Frontier safety table, CBRN row, PDF p. 12 (printed p. 11)

    Note: CBRN evaluations reported at policy level: the uplift TCL is stated as ruled out for chemical, radiological and nuclear; the biological uplift level 1 CCL is stated as not ruled out. Evaluation details in the row are not recorded (hazard rule).

    Inspect sources and checks

Meta

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

    No printed date

    Not found in the recorded search

    Read
    whole card (Markdown file, all sections).
    Terms searched
    “chemical”, “biolog”, “CBRN”, “weapon”, “uplift”, “hazard”.

    Note: No chemical or biological evaluation was found. The search is recorded below.

    Open the search record

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

    No printed date

    Stated, partly

    • “To assess risks related to proliferation of chemical and biological weapons of the Llama 3 family of models, we performed uplift testing”

      Responsibility & Safety, 'Critical and other risks', item 1 (CBRNE)

    Note: The card reports chemical and biological uplift testing for the Llama 3 family, not Llama 3.3 alone, and states no outcome or conclusion for the domain. Not cross-family checked: the maintainer excluded this cell on chemical and biological evaluations from the cross-family check of 2026-11-03.

    Inspect sources and checks

  3. Code World Model Preparedness Report

    revised 2025-10-08

    Stated

    • “we believe that open-source release of CWM is unlikely to lead to additional risk of catastrophic outcomes related to CB1 or CB2”

      Section 3 Chemical & Biological Evaluation, Summary of results, PDF p. 6

    Note: The report states that chemical and biological evaluations were run and gives a conclusion for the framework outcomes CB1 and CB2; with the overall 'moderate' threshold statement under Q05.

    Inspect sources and checks

  4. Muse Spark 1.3 Evaluation Report

    revised 2026-10-23

    Stated

    • “For Muse Spark 1.3, however, we do not rule out a pre-mitigation “high risk” capability finding.”

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

    Note: The report states chemical and biological capability evaluations and the framework outcome language before mitigations; the residual assessment is under Q05.

    Inspect sources and checks

Microsoft

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

    No printed date

    Not found in the recorded search

    Read
    whole card (Markdown file, all sections).
    Terms searched
    “chemical”, “biolog”, “CBRN”, “weapon”, “hazard”.

    Note: No chemical or biological evaluation was found. The search is recorded below. Not cross-family checked: the maintainer excluded this cell on chemical and biological evaluations from the cross-family check of 2026-11-03.

    Open the search record

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

    No printed date

    Not found in the recorded search

    Read
    whole card (Markdown file, all sections).
    Terms searched
    “chemical”, “biolog”, “CBRN”, “weapon”, “hazard”.

    Note: No chemical or biological evaluation was found. The search is recorded below. Not cross-family checked: the maintainer excluded this cell on chemical and biological evaluations from the cross-family check of 2026-11-03.

    Open the search record

  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
    “chemical”, “biolog”, “CBRN”, “weapon”, “hazard”.

    Note: No chemical or biological evaluation was found. The search is recorded below. Not cross-family checked: the maintainer excluded this cell on chemical and biological evaluations from the cross-family check of 2026-11-03.

    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
    “chemical”, “biolog”, “CBRN”, “weapon”, “radiolog”, “nuclear”.

    Note: No chemical or biological evaluation was found. The search is recorded below.

    Open the search record

OpenAI

  1. GPT-4V(ision) System Card

    dated 2023-09-25

    Stated, partly

    • “In the GPT-4 System Card, we explored additional risk areas of CBRN, weapons development, system interaction, and emergent risky properties such as self-replication.”

      Footnote 2, §2.3, PDF p. 6

    • “GPT-4V inherits the assessment in those areas, but this was not a key focus area as image input does not meaningfully alter the capabilities for these categories.”

      Footnote 2, §2.3, PDF p. 6

    • “the model’s generations here can be inaccurate and error prone, limiting its use for such tasks.”

      §2.3.1 Scientific proficiency, PDF p. 7

    Note: The card states that GPT-4V inherits the GPT-4 System Card’s assessment of CBRN risk areas without restating its outcome, and that image input does not meaningfully alter capability in these categories. Red teamers’ scientific-proficiency testing touched on dangerous chemical synthesis; the card states the outputs were inaccurate and error-prone. No evaluation run on GPT-4V with an outcome category or threshold language is reported. Details of the red-team example are not recorded (hazard rule). Search for the rest of the criterion: §2.2 Evaluations; §2.3 External Red Teaming incl. footnote 2; §2.3.1 Scientific proficiency. Terms searched: “CBRN”, “chemical”, “biolog”, “weapon”, “nuclear”, “radiolog”.

    Inspect sources and checks

  2. o1 system card · 2024

    revised 2024-12-19Cites Preparedness · Beta

    Stated

    • “Chemical and Biological Threat Creation Score for o1: Medium (same as o1-preview)”

      §5.5 Chemical and Biological Threat Creation, PDF p. 19

    • “was ultimately designated as earning a “Low” risk under the Preparedness Framework for RN weapon creation capability.”

      §5.6.1 Structured expert probing campaign – radiological & nuclear, PDF p. 27

    Note: Chemical and biological: Medium. Radiological and nuclear are reported in a separate section, with a Low designation for the post-mitigation model. Evaluation methods, tasks and results are not recorded (hazard rule).

    Inspect sources and checks

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

    dated 2025-12-18Cites Preparedness · Version 2

    Stated

    • “As we did for other models in the GPT-5 series, we are treating GPT-5.2-Codex as High risk in the Biological and Chemical domain, and continuing to apply the corresponding safeguards.”

      §5.1.1 Biological and Chemical, PDF p. 9 (printed p. 8)

    Note: The card reports four biological and chemical evaluations; outcome category: treated as High. Evaluation contents and results are not recorded (hazard rule).

    Inspect sources and checks

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

    dated 2026-10-29Cites Preparedness · Version 2

    Stated

    • “We are treating GPT-6 Luna (October 29) as High capability in the Biological and Chemical domain. Below, we report updated results on our High and Critical capability evaluations.”

      §8.1.1 Biological and Chemical Capabilities, PDF p. 13 (printed p. 12)

    Note: Outcome category: High. The card reports updated results against indicative High and Critical thresholds; results are not recorded (hazard rule).

    Inspect sources and checks

xAI

  1. Grok 4.1 Model Card

    dated 2025-11-17

    Stated

    • “Grok 4.1 achieves broadly similar results to Grok 4 and other frontier models on chemical and biological capabilities tests.”

      §2.3.2 Results, p. 4

    Note: Biology and chemistry dual-use evaluations reported, with a stated comparative conclusion and comparisons with human baselines; no threshold language. The section's scope sentence names radiological and nuclear weapons, but the evaluations listed cover biology, chemistry and cyber only. Scores are not recorded (hazard rule). Not cross-family checked: the maintainer excluded this cell on chemical and biological evaluations from the cross-family check of 2026-11-03.

    Inspect sources and checks

  2. Model Card: Grok 4.7

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

    Stated

    • “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: Biological and chemical capability evaluations are reported as FAIF safety-threshold evaluations, run without production safeguards, with the threshold language quoted. Radiological and nuclear items appear only in a refusal evaluation (§9.3). Scores are not recorded (hazard rule).

    Inspect sources and checks