Consensus score
81.5
95% interval 73.8 to 87.4
81.5
95% interval 73.8 to 87.4
67th
IQR spread 12.2 pct
14
Across 4 domains
2
Newest observation 2d ago
Median within-board percentile by capability domain across boards with sufficient peer overlap.
Best verified run per board, with confidence intervals and Pareto frontier status.
| Benchmark | Rank | Score | Cost | Status | Observed |
|---|---|---|---|---|---|
| Legal Agent Benchmark (HLAB-Vals) Vals AI benchmarks · max | #2 / 38Tied #1 | 25%[17.8% to 32.2%] | $0.12 | On frontier. Evidence grade A of A to D: strongest. | 15 Nov 2026 |
| Legal Research Vals AI benchmarks · max | #7 / 38Tied #1 | 49%[42.2% to 55.8%] | $0.06 | On frontier. Evidence grade A of A to D: strongest. | 15 Nov 2026 |
| EMB Vals AI benchmarks · max | #12 / 38Tied #1 | 72.2%[67.6% to 76.7%] | $0.09 | On frontier. Evidence grade A of A to D: strongest. | 15 Nov 2026 |
| Tax Agent Vals AI benchmarks · max | #12 / 38 | 38.2%[31.4% to 45.1%] | $0.03 | On frontier. Evidence grade A of A to D: strongest. | 15 Nov 2026 |
| Terminal-Bench 4 Vals AI benchmarks · max | #14 / 38 | 39.4%[34.9% to 43.9%] | $0.22 | On frontier. Evidence grade A of A to D: strongest. | 15 Nov 2026 |
| Vals Index Vals AI benchmarks · max | #15 / 38 | 60.5%[58.4% to 62.6%] | $0.15 | On frontier. Evidence grade A of A to D: strongest. | 15 Nov 2026 |
| Finance Agent Vals AI benchmarks · max | #16 / 38 | 63.8%[59.9% to 67.8%] | $0.06 | On frontier. Evidence grade A of A to D: strongest. | 15 Nov 2026 |
| Vibe Code Bench Vals AI benchmarks · max · OpenHands | #18 / 38 | 83.7%[78.3% to 89.1%] | $0.05 | On frontier. Evidence grade A of A to D: strongest. | 15 Nov 2026 |
| Code Migration Vals AI benchmarks · max | #24 / 38 | 47.1%[39% to 55.2%] | $0.60 | On frontier. Evidence grade A of A to D: strongest. | 15 Nov 2026 |
| LMArena Agent LMArena · Overall · max | #24 / 64 | 0[0 to 0] | $0.02 | On frontier. Evidence grade A of A to D: strongest. | 15 Nov 2026 |
| LMArena WebDev LMArena · Webdev HTML · max | #30 / 155 | 1,621[1,577 to 1,665] | Unknown | Standard | 15 Nov 2026 |
| LMArena WebDev LMArena · Overall · max | #37 / 156 | 1,608[1,591 to 1,625] | Unknown | Standard | 15 Nov 2026 |
| LMArena WebDev LMArena · Webdev · max | #37 / 156 | 1,608[1,591 to 1,625] | Unknown | Standard | 15 Nov 2026 |
| LMArena WebDev LMArena · Webdev react · max | #38 / 140 | 1,616[1,595 to 1,638] | Unknown | Standard | 15 Nov 2026 |
Direct pairwise win rates for Muse Ember across shared benchmark boards.
11 shared boards
100% win
Head-to-head11 shared boards
55% win
Head-to-head11 shared boards
27% win
Head-to-head11 shared boards
59% win
Head-to-head11 shared boards
82% win
Head-to-head11 shared boards
36% win
Head-to-head11 shared boards
100% win
Head-to-head11 shared boards
45% win
Head-to-head11 shared boards
36% win
Head-to-head11 shared boards
82% win
Head-to-head