B.O.B. Leaderboard
Meta AIreleased 2 Sep 2026
Muse Spark 1.3
#15 of 9523 of 46 benchmarksConsistent
71.2
B.O.B.
Category profile
- Real use (hard to game)2×74
- Reasoning & knowledge60
- Math69
- Coding64
- Agents & real work77
- Independent indexes80
Reality check
Lab tests
68
Real use (6 tests)
74
Lab tests and real-use tests agree. It does 7 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)
$1.25 in · $4.25 out
Cost per task on the AA index
$1.60
Output speed (Artificial Analysis)
128 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 74 | |||
| LMArena WebDevas muse-spark-1.3-max | 1659 | 89%of 82 | LMArena (blind human votes) |
| LMArena Textas muse-spark-1.3-max | 1494 | 89%of 80 | LMArena (blind human votes) |
| EQ-Bench Creative Writingas muse-spark-1.3 | 1906 | 73%of 46 | Sam Paech (open source) |
| BridgeBench (overall)as Muse Spark 1.3 | 606 | 71%of 29 | BridgeMind |
| LMArena Agent Arenaas Muse Spark 1.3 (Max) | 0.035 | 71%of 50 | LMArena (real agent sessions) |
| LiveBench 2.0 (real-work suites)as muse-spark-1.3 | 71.0% | 51%of 39 | LiveBench / Abacus.AI |
| Reasoning & knowledge · Reasoning 60 | |||
| CritPt (physics research)as muse-spark-1.3_xhigh | 26.0% | 82%of 89 | Epoch AI Benchmarking Hub |
| Chess Puzzlesas muse-spark-1.3_max | 37.0% | 56%of 73 | Epoch AI Benchmarking Hub |
| Mystery Game Puzzlesas muse-spark-1.3_max | 25.0% | 44%of 57 | Epoch AI Benchmarking Hub |
| Math · Math 69 | |||
| OTIS Mock AIMEas muse-spark-1.3_xhigh | 99.2% | 83%of 73 | Epoch AI Benchmarking Hub |
| ProofBenchas muse-spark-1.3_max | 58.0% | 67%of 68 | Epoch AI Benchmarking Hub |
| FrontierMath Tiers 1-3as muse-spark-1.3_xhigh | 74.4% | 66%of 60 | Epoch AI Benchmarking Hub |
| FrontierMath Tier 4as muse-spark-1.3_max | 46.3% | 59%of 52 | Epoch AI Benchmarking Hub |
| Coding · Coding 64 | |||
| SciCodeas muse-spark-1.3_xhigh | 59.7% | 91%of 86 | Epoch AI Benchmarking Hub |
| ALE-Benchas muse-spark-1.3_xhigh | 1688.3% | 84%of 71 | Epoch AI Benchmarking Hub |
| CursorBenchas muse-spark-1.3_max | 41.6% | 44%of 17 | Epoch AI Benchmarking Hub |
| FrontierSWEas muse-spark-1.3_max | 25.8% | 37%of 20 | Epoch AI Benchmarking Hub |
| Agents & real work · Agents 77 | |||
| DTBenchas muse-spark-1.3_unknown | 96.5% | 86%of 67 | Epoch AI Benchmarking Hub |
| LMCAas muse-spark-1.3_unknown | 53.9% | 79%of 67 | Epoch AI Benchmarking Hub |
| GDP.pdfas muse-spark-1.3_unknown | 27.6% | 75%of 37 | Epoch AI Benchmarking Hub |
| APEX-Agentsas muse-spark-1.3_unknown | 57.8% | 70%of 48 | Epoch AI Benchmarking Hub |
| Independent indexes · Indexes 80 | |||
| Artificial Analysis Intelligence Indexas Muse Spark 1.3 (max) | 48 | 85%of 54 | Artificial Analysis |
| Epoch Capabilities Indexas Muse Spark 1.3 | 156.8 | 74%of 81 | Epoch AI |