‹ PublicAI Index
The LLM benchmark aggregator.
Claude Fable 5
Anthropic
Strongest in Core abilities (#1 of 204, on 2 of its 3 boards), weakest in Reasoning (#13 of 140). Above par in 24 of 24 scopes. Among the models it meets almost everywhere, it finishes behind Claude Opus 5.5 and ahead of Claude Opus 5 and Claude Fable 5.1.
§ 1 · Profile
What it is good at
Bars run from 50 — the average of the models each source lists — so right of the line is above par. The middle column is the gap to whoever leads that scope.
Core abilities67.4leads#1/2042/3
Language71.5leads#1/571/1
General intelligence69.9−0.2#2/2042/3
Data analysis61.9−4.2#4/571/1
Instruction following63.9−9.8#6/571/1
Professional61.6leads#1/1681/1
Medical63.5−3.3#2/1401/1
Legal62.7−4.4#4/1511/1
Finance61.5−1.4#4/1511/1
Agents67.2−0.9#3/2684/5
Computer use71.7leads#1/371/1
Web research64.2leads#1/41/1
Knowledge work67.7−5.9#11/1782/2
Human preference67−0.5#3/3421/1
Human preference67−0.5#3/3421/1
Coding65.4−3.7#3/1653/5
Code generation67.6−2.4#3/771/2
Agentic coding64.8−3.9#4/1573/4
Knowledge60.7−2.8#4/1381/2
Academic knowledge62.8−1.1#3/1231/1
Reasoning62.9−4#5/1783/4
Mathematics63.8−1.9#5/1402/2
Science61.9−1.6#11/1221/1
Reasoning62.3−6.8#13/1402/3
§ 2 · Head to head
What it beats, and what beats it
The same models turn up scope after scope. Counted once: where both were placed, who finished higher. Bars run right for Claude Fable 5, left for the other.
§ 3 · Sources
Where the numbers come from
9 publications, 31 figures. Every one links to the page it was read from.
LMArena Text 1504
GDPval-AA 1595
AA-Briefcase 1541
Terminal-Bench 44.5% (Claude Code)
ARC-AGI-2 89.2%
LiveBench 83LiveBench · Reasoning 89.7LiveBench · Coding 86LiveBench · Agentic Coding 62.2LiveBench · Mathematics 96LiveBench · Data Analysis 80.5LiveBench · Language 90.7LiveBench · Instruction Following 75.8
OSWorld 86%
Kagi LLM Benchmark 91.4%
Vals · Legal Research Bench 49.52%Vals · LegalBench 88.56%Vals · Harvey Legal Agent Benchmark 11.25%Vals · Finance Agent 56.31%Vals · CorpFin 71.83%Vals · TaxEval 76.94%Vals · MortgageTax 68.92%Vals · MedCode 56.07%Vals · MedScribe 88.52%Vals · Web Search Index 48.45% (Exa)Vals · SWE-bench Verified 95% (Mini-SWE-agent)Vals · Vibe Code Bench 90.35% (OpenHands)Vals · Code Migration 55.06%Vals · GPQA Diamond 93.18%Vals · MMLU Pro 91.5%Vals · ProofBench 95%
Badge
[](https://publicai.io/model-index/m/claude-fable-5)