B.O.B. Leaderboard
Anthropicreleased 1 Sep 2026
Claude Fable 5.1
#3 of 9533 of 46 benchmarksConsistent
91.6
B.O.B.
Category profile
- Real use (hard to game)2×95
- Reasoning & knowledge89
- Math94
- Coding92
- Agents & real work82
- Independent indexes94
Reality check
Lab tests
89
Real use (7 tests)
95
Lab tests and real-use tests agree. It does 6 points better on tests that are hard to train for than on the usual lab tests.
What counts as real usePrice per 1M tokens (OpenRouter)
$10.00 in · $50.00 out
Cost per task on the AA index
$7.63
Output speed (Artificial Analysis)
71 tok/s
§Receipts
Every benchmark result
“Beats” = the share of tracked models this one beat on that test. The raw score is its best reported setting.
| Benchmark | Raw | Beats | Source |
|---|---|---|---|
| Real use (hard to game) · Real use 95 | |||
| LiveBench 2.0 (real-work suites)as claude-fable-5.1 | 88.8% | 100%of 39 | LiveBench / Abacus.AI |
| EQ-Bench Creative Writingas claude-fable-5-1 | 2162 | 98%of 46 | Sam Paech (open source) |
| LMArena Agent Arenaas Claude Fable 5.1 (Max) | 0.143 | 96%of 50 | LMArena (real agent sessions) |
| LMArena WebDevas claude-fable-5.1-max | 1743 | 95%of 82 | LMArena (blind human votes) |
| BridgeBench (overall)as Claude Fable 5.1 | 689 | 93%of 29 | BridgeMind |
| Vals Indexas anthropic/claude-fable-5-1 | 65.8% | 93%of 42 | Vals AI (private test sets) |
| LMArena Textas claude-fable-5.1-max | 1501 | 92%of 80 | LMArena (blind human votes) |
| Reasoning & knowledge · Reasoning 89 | |||
| CritPt (physics research)as claude-fable-5-1_xhigh | 31.1% | 94%of 89 | Epoch AI Benchmarking Hub |
| Humanity's Last Examas claude-fable-5-1_xhigh | 46.5% | 92%of 13 | Epoch AI Benchmarking Hub |
| SimpleQA Verifiedas claude-fable-5-1_max | 70.8% | 91%of 57 | Epoch AI Benchmarking Hub |
| Blueprint-Bench 2as claude-fable-5-1_unknown | 41.9% | 90%of 31 | Epoch AI Benchmarking Hub |
| ARC-AGI-2as claude-fable-5-1_max | 90.0% | 88%of 44 | Epoch AI Benchmarking Hub |
| Mystery Game Puzzlesas claude-fable-5-1_max | 58.0% | 88%of 57 | Epoch AI Benchmarking Hub |
| Chess Puzzlesas claude-fable-5-1_max | 48.0% | 78%of 73 | Epoch AI Benchmarking Hub |
| Math · Math 94 | |||
| ProofBenchas claude-fable-5-1_max | 100.0% | 99%of 68 | Epoch AI Benchmarking Hub |
| FrontierMath Tiers 1-3as claude-fable-5-1_max | 90.2% | 95%of 60 | Epoch AI Benchmarking Hub |
| OTIS Mock AIMEas claude-fable-5-1_max | 100.0% | 94%of 73 | Epoch AI Benchmarking Hub |
| FrontierMath Tier 4as claude-fable-5-1_max | 87.8% | 90%of 52 | Epoch AI Benchmarking Hub |
| Coding · Coding 92 | |||
| GSO (code optimisation)as claude-fable-5-1_unknown | 88.2% | 100%of 11 | Epoch AI Benchmarking Hub |
| SciCodeas claude-fable-5-1_max | 63.1% | 99%of 86 | Epoch AI Benchmarking Hub |
| WeirdMLas claude-fable-5-1_max | 92.9% | 96%of 52 | Epoch AI Benchmarking Hub |
| ALE-Benchas claude-fable-5-1_high | 2143.2% | 91%of 71 | Epoch AI Benchmarking Hub |
| FrontierCodeas claude-fable-5-1_medium | 50.9% | 89%of 45 | Epoch AI Benchmarking Hub |
| FrontierSWEas claude-fable-5-1_max | 56.3% | 84%of 20 | Epoch AI Benchmarking Hub |
| CursorBenchas claude-fable-5-1_max | 51.8% | 81%of 17 | Epoch AI Benchmarking Hub |
| Agents & real work · Agents 82 | |||
| LMCAas claude-fable-5-1_unknown | 65.5% | 99%of 67 | Epoch AI Benchmarking Hub |
| DTBenchas claude-fable-5-1_unknown | 97.6% | 96%of 67 | Epoch AI Benchmarking Hub |
| APEX-Agentsas claude-fable-5-1_unknown | 68.6% | 94%of 48 | Epoch AI Benchmarking Hub |
| GDP.pdfas claude-fable-5-1_high | 29.6% | 81%of 37 | Epoch AI Benchmarking Hub |
| Remote Labor Indexas claude-fable-5-1_unknown | 17.9% | 75%of 9 | Epoch AI Benchmarking Hub |
| Vending-Bench 2as claude-fable-5-1_unknown | 5421.6% | 46%of 47 | Epoch AI Benchmarking Hub |
| Independent indexes · Indexes 94 | |||
| Epoch Capabilities Indexas Claude Fable 5.1 | 164.7 | 96%of 81 | Epoch AI |
| Artificial Analysis Intelligence Indexas Claude Fable 5.1 (max with fallback) | 53 | 93%of 54 | Artificial Analysis |