Consensus score
66.6
95% interval 58.7 to 71.5
66.6
95% interval 58.7 to 71.5
43th
IQR spread 41.0 pct
69
Across 6 domains
3
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 |
|---|---|---|---|---|---|
| Code Migration Vals AI benchmarks · max | #18 / 38 | 52%[45.4% to 58.6%] | $11.81 | Standard | 15 Nov 2026 |
| Vibe Code Bench Vals AI benchmarks · max · OpenHands | #20 / 38 | 82.3%[76.1% to 88.5%] | $0.36 | Standard | 15 Nov 2026 |
| Legal Agent Benchmark (HLAB-Vals) Vals AI benchmarks · max | #22 / 38 | 10.8%[5.5% to 16.2%] | $0.80 | Standard | 15 Nov 2026 |
| Legal Research Vals AI benchmarks · max | #22 / 38 | 40.9%[34.2% to 47.6%] | $1.09 | Standard | 15 Nov 2026 |
| Tax Agent Vals AI benchmarks · max | #27 / 38 | 26.8%[20.7% to 32.8%] | $0.73 | Standard | 15 Nov 2026 |
| Terminal-Bench 4 Vals AI benchmarks · max | #29 / 38 | 14.1%[10.2% to 18.1%] | $3.31 | Standard | 15 Nov 2026 |
| Vals Index Vals AI benchmarks · max | #29 / 38 | 53%[50.9% to 55%] | $2.50 | Standard | 15 Nov 2026 |
| Finance Agent Vals AI benchmarks · max | #31 / 38 | 57.1%[53% to 61.2%] | $0.82 | Standard | 15 Nov 2026 |
| EMB Vals AI benchmarks · max | #32 / 38 | 65.1%[60.5% to 69.7%] | $1.08 | Standard | 15 Nov 2026 |
| FrontierCode Epoch AI Benchmarking Hub · high · devin-cli | #34 / 51 | 28.6% | Unknown | Standard | Not stated |
| GPQA diamond Epoch AI Benchmarking Hub · max | #39 / 338 | 91.7%[88.7% to 94.7%] | Unknown | Standard | 18 Aug 2026 |
| LMArena Text (style control) LMArena · Industry legal and government · high | #40 / 393 | 1,482[1,461 to 1,502] | Unknown | Standard | 11 Nov 2026 |
| LMArena WebDev LMArena · Webdev react · high | #43 / 140 | 1,586[1,574 to 1,598] | Unknown | Standard | 15 Nov 2026 |
| LMArena WebDev LMArena · Overall · high | #45 / 156 | 1,581[1,571 to 1,590] | Unknown | Standard | 15 Nov 2026 |
| LMArena WebDev LMArena · Webdev · high | #45 / 156 | 1,581[1,571 to 1,590] | Unknown | Standard | 15 Nov 2026 |
| LMArena WebDev LMArena · Webdev HTML · high | #45 / 155 | 1,561[1,534 to 1,589] | Unknown | Standard | 15 Nov 2026 |
| LMArena Text (style control) LMArena · English · high | #46 / 421 | 1,478[1,469 to 1,488] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Industry business and management and financial operations · high | #49 / 414 | 1,467[1,453 to 1,481] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Industry life and physical and social science · high | #50 / 419 | 1,487[1,471 to 1,502] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Industry writing and literature and language · high | #52 / 420 | 1,453[1,441 to 1,465] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · English · high | #54 / 421 | 1,458[1,448 to 1,467] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Industry legal and government · high | #54 / 393 | 1,460[1,439 to 1,480] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Industry writing and literature and language · high | #54 / 420 | 1,442[1,430 to 1,454] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Creative writing · high | #54 / 419 | 1,444[1,430 to 1,458] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Instruction following · high | #54 / 421 | 1,460[1,450 to 1,471] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Creative writing · high | #55 / 419 | 1,437[1,423 to 1,451] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Instruction following · high | #55 / 421 | 1,448[1,437 to 1,458] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Industry business and management and financial operations · high | #56 / 414 | 1,442[1,428 to 1,456] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Industry entertainment and sports and media · high | #58 / 419 | 1,424[1,412 to 1,437] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Longer query · high | #58 / 399 | 1,457[1,448 to 1,466] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Hard prompts English · high | #59 / 419 | 1,489[1,477 to 1,501] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Longer query · high | #59 / 399 | 1,474[1,465 to 1,484] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Exclude ties · high | #60 / 421 | 1,468[1,460 to 1,477] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Overall · high | #60 / 421 | 1,464[1,457 to 1,471] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Hard prompts English · high | #62 / 419 | 1,463[1,451 to 1,475] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Industry entertainment and sports and media · high | #62 / 419 | 1,436[1,423 to 1,448] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Hard prompts · high | #64 / 421 | 1,482[1,474 to 1,490] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Hard prompts · high | #65 / 421 | 1,457[1,449 to 1,464] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Overall · high | #65 / 421 | 1,445[1,438 to 1,451] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Non English · high | #65 / 421 | 1,446[1,438 to 1,454] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Exclude ties · high | #66 / 421 | 1,441[1,432 to 1,450] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Industry life and physical and social science · high | #66 / 419 | 1,457[1,442 to 1,473] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Russian · high | #66 / 385 | 1,444[1,427 to 1,461] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Industry medicine and healthcare · high | #66 / 391 | 1,477[1,454 to 1,501] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Multi turn · high | #67 / 419 | 1,452[1,436 to 1,467] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Multi turn · high | #67 / 419 | 1,469[1,453 to 1,484] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Spanish · high | #67 / 304 | 1,456[1,423 to 1,488] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Russian · high | #69 / 385 | 1,456[1,439 to 1,473] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Coding · high | #74 / 416 | 1,503[1,492 to 1,515] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Industry software and IT services · high | #74 / 421 | 1,491[1,481 to 1,501] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Coding · high | #76 / 416 | 1,467[1,456 to 1,479] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Non English · high | #78 / 421 | 1,429[1,421 to 1,437] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Expert · high | #78 / 371 | 1,481[1,462 to 1,500] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Chinese · high | #79 / 395 | 1,478[1,455 to 1,501] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Expert · high | #81 / 371 | 1,455[1,436 to 1,473] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Chinese · high | #81 / 395 | 1,492[1,469 to 1,514] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Math · high | #81 / 404 | 1,449[1,421 to 1,477] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Industry mathematical · high | #83 / 402 | 1,455[1,429 to 1,481] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Industry software and IT services · high | #84 / 421 | 1,460[1,450 to 1,470] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Industry medicine and healthcare · high | #86 / 391 | 1,446[1,423 to 1,469] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Spanish · high | #89 / 304 | 1,436[1,404 to 1,469] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · French · high | #89 / 301 | 1,466[1,433 to 1,500] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Industry mathematical · high | #95 / 402 | 1,438[1,412 to 1,465] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Math · high | #95 / 404 | 1,433[1,405 to 1,461] | Unknown | Standard | 11 Nov 2026 |
| ARC-AGI-2 Epoch AI Benchmarking Hub · max | #98 / 299 | 61.3% | Unknown | Standard | Not stated |
| LMArena Text LMArena · French · high | #101 / 301 | 1,440[1,407 to 1,474] | Unknown | Standard | 11 Nov 2026 |
| ARC-AGI-2 Epoch AI Benchmarking Hub · high | #105 / 299 | 59.7% | Unknown | Standard | Not stated |
| ARC-AGI-2 Epoch AI Benchmarking Hub · low | #114 / 299 | 56.3% | Unknown | Standard | Not stated |
| ARC-AGI-2 Epoch AI Benchmarking Hub · none | #269 / 299 | 0.8% | Unknown | Standard | Not stated |
Same benchmark and harness measured across multiple reasoning effort settings.
Epoch AI Benchmarking Hub
0.8
56.3
59.7
61.3
Direct pairwise win rates for DeepSeek V4 Pro 0813 across shared benchmark boards.
15 shared boards
33% win
Head-to-head15 shared boards
27% win
Head-to-head15 shared boards
87% win
Head-to-head15 shared boards
27% win
Head-to-head15 shared boards
20% win
Head-to-head15 shared boards
0% win
Head-to-head15 shared boards
20% win
Head-to-head15 shared boards
37% win
Head-to-head15 shared boards
100% win
Head-to-head15 shared boards
20% win
Head-to-head