Clinical Benchmarks

Claude Fable 5.2

Released 9 Nov 2026proprietaryAlso written as Claude Fable 5.2, claude-fable-5-2, anthropic/claude-fable-5-2

4 results on 4 of 14 boards, latest measured Nov 2026. Compare with other models

Clinical Benchmarks Index64.5Rank 1 of 1314 of 12 boardsThis site’s own calculation. Show the working How it is computed

On the Clinical Benchmarks Index

The top 15 on the index.The top 10 on the index. Select a column to open that model.

Board by board

Leading models on each board, with Claude Fable 5.2 included, on that board's own scale. Its own column is in colour.

Every result, with its source

Each line sits on its own board's scale. Tall tick: the board leader.

Clinical reasoning and knowledge

  1. HealthBench Professional
    Artificial Analysis run at Anthropic request; max effort; default fallback; GPT-5.4 low grader; length-adjusted; raw 71.1%
    0.677
    Rank 3 of 46Leader GPT-6 Astra (Anthropic run) 0.703
    Independent run
    Measured Nov 2026

Documentation and coding

  1. MLCR-AA
    overall pass rate; Max, Default Fallback
    43.9
    Rank 1 of 30 here, 27 models on the boardLeads this board
    Independent run
    Measured Nov 2026

EHR and workflow agents

  1. PhysicianBench
    Anthropic harness; 100 tasks, five attempts each; max effort; fallbacks on; Claude Opus 5 grader
    64.4
    Rank 2 of 29 here, 12 models on the boardLeader Claude Opus 5.5 (max) 68.4
    Vendor-reported
    Measured Nov 2026
  2. DAYJOB Health
    Score (pass@1 at 95%), reasoning adaptive/max
    45.0
    Rank 2 of 18 here, 15 models on the boardLeader Claude Opus 5.5 48.3
    Official leaderboard
    Measured Nov 2026

Index working

Show the working: each board’s result, how hard the board runs and the field average

The Clinical Benchmarks Index is this site's own calculation from 12 boards. How it is computed

  • HealthBench Professional0.67767.7+5.0 easy board62.70.743 weightArtificial Analysis run at Anthropic request; max effort; default fallback; GPT-5.4 low grader; length-adjusted; raw 71.1%
  • PhysicianBench64.464.4−15.7 hard board80.10.79 weightAnthropic harness; 100 tasks, five attempts each; max effort; fallbacks on; Claude Opus 5 grader
  • DAYJOB Health45.045.0−29.6 hard board74.60.6 weightScore (pass@1 at 95%), reasoning adaptive/max
  • MLCR-AA43.943.9−25.9 hard board69.80.6 weightoverall pass rate; Max, Default Fallback
Adjusted average of 4 boards, by weight
71.9
Field average, counts for 27.3% with 4 of 12 boards
44.9
Index
= 64.5

Score: the result on the board’s own 0 to 100 scale (dot; tick: the board’s leader). Board runs: how far the board’s scores sit above or below the same models’ scores on the other boards. Adjusted: the score minus that offset. The index is the adjusted average with the field average mixed in, more of it the fewer boards a model has. Each result links to its quote and page; the line under it is the setting it was measured with.

Sources

The quote, page and document for each of the 4 results
  1. 3HealthBench Professional Artificial Analysis run… 0.677
    Printed as 67.7%Independent run, measured Nov 2026Configuration: Artificial Analysis run at Anthropic request; max effort; default fallback; GPT-5.4 low grader; length-adjusted; raw 71.1%
    System Card: Claude Fable 5.2 and Claude Mythos 5.2 system card, Anthropic, 9 Nov 2026. PDF p. 246, section 8.16.2
    Claude Fable 5.2 scored 71.1% (67.7% adjusted)
    Every result from this document
  2. 2PhysicianBench Anthropic harness 64.4
    Printed as 64.4%Vendor-reported, measured Nov 2026Configuration: Anthropic harness; 100 tasks, five attempts each; max effort; fallbacks on; Claude Opus 5 grader
    System Card: Claude Fable 5.2 and Claude Mythos 5.2 system card, Anthropic, 9 Nov 2026. PDF p. 247, section 8.16.3
    Claude Fable 5.2 passed 64.4% of attempts.
    Every result from this document
  3. 2DAYJOB Health Score 45.0
    Printed as 45.0%Official leaderboard, measured Nov 2026Configuration: Score (pass@1 at 95%), reasoning adaptive/max
    DAYJOB: Healthcare on the Fireworks Specialized Intelligence Index official leaderboard, Fireworks AI, 28 Oct 2026. leaderboard row 3, read 2026-11-21
    3 Claude Fable 5.2 Anthropic 45.0% $28.65
    Every result from this document
  4. 1MLCR-AA overall pass rate 43.9
    Printed as 43.9%Independent run, measured Nov 2026Configuration: overall pass rate; Max, Default Fallback
    MLCR-AA (Medical Long Context Reasoning), Artificial Analysis official leaderboard, Artificial Analysis, 5 Nov 2026. MLCR-AA pass rate highlight
    Claude Fable 5.2 (Max, Default Fallback) has the highest MLCR-AA score, with a pass rate of 43.9%.
    Every result from this document

Other Anthropic models: Claude Fable 5, Claude Fable 5.1, Claude Haiku 4.5, Claude Haiku 5.5, Claude Opus 4.1, Claude Opus 4.5, Claude Opus 4.6, Claude Opus 4.7, Claude Opus 4.8, Claude Opus 5, Claude Opus 5.5, Claude Sonnet 4, Claude Sonnet 4.5, Claude Sonnet 4.6, Claude Sonnet 5, Claude Sonnet 5.5