B.O.B. Leaderboard
Anthropicreleased 9 Jun 2026
Claude Fable 5
#9 of 9537 of 46 benchmarksConsistent
86.2
B.O.B.
Category profile
- Real use (hard to game)2×89
- Reasoning & knowledge77
- Math91
- Coding87
- Agents & real work81
- Independent indexes90
Reality check
Lab tests
84
Real use (10 tests)
89
Lab tests and real-use tests agree. It does 5 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
—
Output speed (Artificial Analysis)
—
§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 89 | |||
| SimpleBenchas claude-fable-5_max | 81.9% | 100%of 30 | Epoch AI Benchmarking Hub |
| SWE-rebenchas Fable 5 [high] | 64.5% | 100%of 34 | Nebius |
| Kagi LLM Benchmarkas claude-fable-5-thinking | 91.4% | 100%of 22 | Kagi |
| LMArena Textas claude-fable-5-high | 1503 | 96%of 80 | LMArena (blind human votes) |
| LiveBench 2.0 (real-work suites)as claude-fable-5 | 83.0% | 87%of 39 | LiveBench / Abacus.AI |
| LMArena Agent Arenaas Claude Fable 5 (High) | 0.085 | 84%of 50 | LMArena (real agent sessions) |
| Vals Indexas anthropic/claude-fable-5 | 61.4% | 83%of 42 | Vals AI (private test sets) |
| EQ-Bench Creative Writingas claude-fable-5 | 1943 | 80%of 46 | Sam Paech (open source) |
| LMArena WebDevas claude-fable-5-high | 1624 | 79%of 82 | LMArena (blind human votes) |
| BridgeBench (overall)as Claude Fable 5 | 628 | 79%of 29 | BridgeMind |
| Reasoning & knowledge · Reasoning 77 | |||
| EnigmaEvalas claude-fable-5_high | 39.3% | 100%of 10 | Epoch AI Benchmarking Hub |
| SimpleQA Verifiedas claude-fable-5_xhigh | 70.7% | 89%of 57 | Epoch AI Benchmarking Hub |
| CritPt (physics research)as claude-fable-5_max | 28.6% | 88%of 89 | Epoch AI Benchmarking Hub |
| ARC-AGI-2as claude-fable-5_max | 89.2% | 83%of 44 | Epoch AI Benchmarking Hub |
| Mystery Game Puzzlesas claude-fable-5_max | 52.0% | 82%of 57 | Epoch AI Benchmarking Hub |
| Blueprint-Bench 2as claude-fable-5_unknown | 38.6% | 82%of 31 | Epoch AI Benchmarking Hub |
| Chess Puzzlesas claude-fable-5_max | 47.0% | 74%of 73 | Epoch AI Benchmarking Hub |
| GPQA Diamondas claude-fable-5_max | 85.9% | 21%of 70 | Epoch AI Benchmarking Hub |
| Math · Math 91 | |||
| FrontierMath Tier 4as claude-fable-5_max | 90.2% | 94%of 52 | Epoch AI Benchmarking Hub |
| OTIS Mock AIMEas claude-fable-5_high | 100.0% | 94%of 73 | Epoch AI Benchmarking Hub |
| ProofBenchas claude-fable-5_max | 95.0% | 90%of 68 | Epoch AI Benchmarking Hub |
| FrontierMath Tiers 1-3as claude-fable-5_max | 87.0% | 86%of 60 | Epoch AI Benchmarking Hub |
| Coding · Coding 87 | |||
| FrontierCodeas claude-fable-5_unknown | 53.5% | 98%of 45 | Epoch AI Benchmarking Hub |
| SciCodeas claude-fable-5_max | 61.0% | 96%of 86 | Epoch AI Benchmarking Hub |
| WeirdMLas claude-fable-5_max | 91.9% | 94%of 52 | Epoch AI Benchmarking Hub |
| ALE-Benchas claude-fable-5_high | 2041.3% | 90%of 71 | Epoch AI Benchmarking Hub |
| DeepSWEas claude-fable-5_xhigh | 69.9% | 84%of 26 | Epoch AI Benchmarking Hub |
| GSO (code optimisation)as claude-fable-5_unknown | 78.4% | 80%of 11 | Epoch AI Benchmarking Hub |
| FrontierSWEas claude-fable-5_max | 47.0% | 68%of 20 | Epoch AI Benchmarking Hub |
| Agents & real work · Agents 81 | |||
| DTBenchas claude-fable-5_max | 98.4% | 99%of 67 | Epoch AI Benchmarking Hub |
| Game Boy Advance evalas claude-fable-5_unknown | 74.5% | 96%of 23 | Epoch AI Benchmarking Hub |
| LMCAas claude-fable-5_high | 61.1% | 94%of 67 | Epoch AI Benchmarking Hub |
| GDP.pdfas claude-fable-5_unknown | 30.0% | 83%of 37 | Epoch AI Benchmarking Hub |
| APEX-Agentsas claude-fable-5 | 63.6% | 81%of 48 | Epoch AI Benchmarking Hub |
| Remote Labor Indexas claude-fable-5 | 16.1% | 63%of 9 | Epoch AI Benchmarking Hub |
| Vending-Bench 2as claude-fable-5_high | 5680.3% | 50%of 47 | Epoch AI Benchmarking Hub |
| Independent indexes · Indexes 90 | |||
| Epoch Capabilities Indexas Claude Fable 5 | 162.1 | 90%of 81 | Epoch AI |