Consensus score
92.4
95% interval 84.8 to 99.0
92.4
95% interval 84.8 to 99.0
92th
IQR spread 10.8 pct
111
Across 8 domains
4
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 |
|---|---|---|---|---|---|
| FrontierCode Epoch AI Benchmarking Hub · medium · Claude Code | #1 / 51Tied #1 | 54.6% | Unknown | Standard | Not stated |
| LMArena Text LMArena · Korean · high | #1 / 289Tied #1 | 1,505[1,465 to 1,544] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Korean · high | #1 / 289Tied #1 | 1,492[1,452 to 1,531] | Unknown | Standard | 11 Nov 2026 |
| LMArena Vision LMArena · Diagram · high | #1 / 123Tied #1 | 1,354[1,335 to 1,372] | Unknown | Standard | 9 Nov 2026 |
| LMArena WebDev LMArena · Image to webdev · max | #1 / 64Tied #1 | 1,756[1,738 to 1,775] | Unknown | Standard | 15 Nov 2026 |
| Terminal-Bench 4 Vals AI benchmarks · max | #1 / 38Tied #1 | 65.2%[65.2% to 65.2%] | $12.04 | On frontier. Evidence grade A of A to D: strongest. | 15 Nov 2026 |
| Finance Agent Vals AI benchmarks · max | #2 / 38Tied #1 | 67.8%[63.8% to 71.9%] | $8.91 | Standard | 15 Nov 2026 |
| LMArena Agent LMArena · Overall · max | #2 / 64 | 0.2[0.2 to 0.2] | $5.65 | On frontier. Evidence grade A of A to D: strongest. | 15 Nov 2026 |
| LMArena Text LMArena · Coding · high | #2 / 416Tied #1 | 1,543[1,533 to 1,554] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Creative writing · high | #2 / 419Tied #1 | 1,517[1,506 to 1,528] | Unknown | Standard | 11 Nov 2026 |
| LMArena Vision LMArena · Homework · high | #2 / 118Tied #1 | 1,357[1,333 to 1,380] | Unknown | Standard | 9 Nov 2026 |
| LMArena Vision LMArena · OCR · high | #2 / 123Tied #1 | 1,332[1,321 to 1,344] | Unknown | Standard | 9 Nov 2026 |
| LMArena Vision LMArena · Overall · high | #2 / 167Tied #1 | 1,324[1,314 to 1,334] | Unknown | Standard | 9 Nov 2026 |
| LMArena WebDev LMArena · Overall · max | #2 / 156Tied #1 | 1,816[1,804 to 1,827] | Unknown | Standard | 15 Nov 2026 |
| LMArena WebDev LMArena · Webdev · max | #2 / 156Tied #1 | 1,816[1,804 to 1,827] | Unknown | Standard | 15 Nov 2026 |
| LMArena WebDev LMArena · Webdev react · max | #2 / 140Tied #1 | 1,871[1,856 to 1,885] | Unknown | Standard | 15 Nov 2026 |
| LMArena Text LMArena · Creative writing · high | #3 / 419Tied #1 | 1,532[1,520 to 1,543] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · English · high | #3 / 421Tied #1 | 1,521[1,513 to 1,530] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Exclude ties · high | #3 / 421 | 1,534[1,526 to 1,542] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Hard prompts · high | #3 / 421 | 1,531[1,524 to 1,538] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Hard prompts English · high | #3 / 419Tied #1 | 1,536[1,525 to 1,547] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Industry entertainment and sports and media · high | #3 / 419Tied #1 | 1,508[1,497 to 1,518] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Industry software and IT services · high | #3 / 421 | 1,533[1,524 to 1,542] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Industry writing and literature and language · high | #3 / 420Tied #1 | 1,528[1,518 to 1,538] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Instruction following · high | #3 / 421Tied #1 | 1,535[1,526 to 1,544] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Japanese · high | #3 / 288Tied #1 | 1,527[1,488 to 1,565] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Longer query · high | #3 / 399 | 1,528[1,519 to 1,536] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Overall · high | #3 / 421 | 1,514[1,508 to 1,520] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Spanish · high | #3 / 304Tied #1 | 1,510[1,484 to 1,536] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Industry entertainment and sports and media · high | #3 / 419 | 1,496[1,486 to 1,507] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Industry writing and literature and language · high | #3 / 420Tied #1 | 1,516[1,506 to 1,526] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Instruction following · high | #3 / 421Tied #1 | 1,518[1,509 to 1,527] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Japanese · high | #3 / 288Tied #1 | 1,524[1,486 to 1,563] | Unknown | Standard | 11 Nov 2026 |
| LMArena Vision (style control) LMArena · Diagram · high | #3 / 123Tied #1 | 1,339[1,320 to 1,357] | Unknown | Standard | 9 Nov 2026 |
| LMArena Vision (style control) LMArena · Homework · high | #3 / 118Tied #1 | 1,345[1,322 to 1,368] | Unknown | Standard | 9 Nov 2026 |
| Code Migration Vals AI benchmarks · max | #4 / 38Tied #1 | 80.5%[72.5% to 88.5%] | $72.30 | Standard | 15 Nov 2026 |
| EMB Vals AI benchmarks · max | #4 / 38Tied #1 | 75.8%[71.1% to 80.6%] | $9.33 | Standard | 15 Nov 2026 |
| LMArena Text LMArena · Expert · high | #4 / 371Tied #1 | 1,555[1,539 to 1,571] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Non English · high | #4 / 421 | 1,503[1,496 to 1,510] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Russian · high | #4 / 385 | 1,517[1,504 to 1,531] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · English · high | #4 / 421Tied #1 | 1,515[1,507 to 1,524] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Exclude ties · high | #4 / 421 | 1,526[1,518 to 1,534] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Expert · high | #4 / 371Tied #1 | 1,545[1,529 to 1,561] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Hard prompts · high | #4 / 421 | 1,530[1,523 to 1,537] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Overall · high | #4 / 421 | 1,507[1,501 to 1,513] | Unknown | Standard | 11 Nov 2026 |
| LMArena Vision LMArena · English · high | #4 / 167Tied #1 | 1,321[1,305 to 1,338] | Unknown | Standard | 9 Nov 2026 |
| Vals Index Vals AI benchmarks · max | #4 / 38Tied #1 | 70.1%[68.2% to 72.1%] | $24.77 | Standard | 15 Nov 2026 |
| ARC-AGI-2 Epoch AI Benchmarking Hub · high | #5 / 299 | 93.3% | Unknown | Standard | Not stated |
| CursorBench CursorBench · max · Cursor | #5 / 91 | 57.8% | $13.43 | Standard | Not stated |
| LMArena Text LMArena · German · high | #5 / 320 | 1,513[1,474 to 1,553] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Industry life and physical and social science · high | #5 / 419 | 1,525[1,513 to 1,538] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Chinese · high | #5 / 395Tied #1 | 1,565[1,547 to 1,582] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Hard prompts English · high | #5 / 419Tied #1 | 1,535[1,524 to 1,546] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Industry mathematical · high | #5 / 402Tied #1 | 1,523[1,503 to 1,544] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Longer query · high | #5 / 399 | 1,520[1,511 to 1,529] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Non English · high | #5 / 421 | 1,494[1,487 to 1,501] | Unknown | Standard | 11 Nov 2026 |
| Legal Research Vals AI benchmarks · max | #5 / 38Tied #1 | 50.5%[43.7% to 57.3%] | $23.61 | Standard | 15 Nov 2026 |
| Tax Agent Vals AI benchmarks · max | #5 / 38Tied #1 | 44.5%[37.5% to 51.6%] | $14.40 | Standard | 15 Nov 2026 |
| Vibe Code Bench Vals AI benchmarks · max · OpenHands | #5 / 38Tied #1 | 90.3%[87.3% to 93.3%] | $35.77 | Standard | 15 Nov 2026 |
| LMArena Text LMArena · Chinese · high | #6 / 395 | 1,573[1,555 to 1,590] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Industry medicine and healthcare · high | #6 / 391Tied #1 | 1,511[1,492 to 1,531] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Math · high | #6 / 404Tied #1 | 1,520[1,498 to 1,542] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Coding · high | #6 / 416Tied #1 | 1,547[1,536 to 1,558] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · German · high | #6 / 320 | 1,510[1,471 to 1,549] | Unknown | Standard | 11 Nov 2026 |
| LMArena Vision LMArena · Chinese · high | #6 / 135Tied #1 | 1,380[1,341 to 1,419] | Unknown | Standard | 9 Nov 2026 |
| LMArena Vision (style control) LMArena · OCR · high | #6 / 123 | 1,312[1,300 to 1,323] | Unknown | Standard | 9 Nov 2026 |
| LMArena Vision (style control) LMArena · Overall · high | #6 / 167 | 1,295[1,285 to 1,305] | Unknown | Standard | 9 Nov 2026 |
| ARC-AGI-2 Epoch AI Benchmarking Hub · xhigh | #7 / 299 | 92.5% | Unknown | Standard | Not stated |
| LMArena Text LMArena · Industry legal and government · high | #7 / 393Tied #1 | 1,519[1,501 to 1,537] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Industry mathematical · high | #7 / 402Tied #1 | 1,532[1,512 to 1,553] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text LMArena · Polish · high | #7 / 243Tied #1 | 1,511[1,471 to 1,551] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Russian · high | #7 / 385 | 1,514[1,500 to 1,527] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Spanish · high | #7 / 304Tied #1 | 1,494[1,468 to 1,521] | Unknown | Standard | 11 Nov 2026 |
| LMArena Vision (style control) LMArena · Chinese · high | #7 / 135Tied #1 | 1,353[1,313 to 1,392] | Unknown | Standard | 9 Nov 2026 |
| LMArena Agent LMArena · Overall · high | #8 / 64 | 0.1[0.1 to 0.1] | $2.12 | Standard | 15 Nov 2026 |
| LMArena Text LMArena · Industry business and management and financial operations · high | #8 / 414 | 1,494[1,482 to 1,506] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Industry life and physical and social science · high | #8 / 419Tied #1 | 1,522[1,509 to 1,535] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Industry software and IT services · high | #8 / 421Tied #1 | 1,534[1,525 to 1,543] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Math · high | #8 / 404Tied #1 | 1,514[1,493 to 1,536] | Unknown | Standard | 11 Nov 2026 |
| LMArena WebDev LMArena · Webdev HTML · max | #8 / 155 | 1,729[1,685 to 1,772] | Unknown | Standard | 15 Nov 2026 |
| CursorBench CursorBench · xhigh · Cursor | #9 / 91 | 56% | $6.98 | Standard | Not stated |
| CursorBench CursorBench · high · Cursor | #9 / 91 | 56% | $3.97 | On frontier. Evidence grade A of A to D: strongest. | Not stated |
| LMArena Text LMArena · Multi turn · high | #10 / 419 | 1,502[1,489 to 1,516] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Industry medicine and healthcare · high | #10 / 391Tied #1 | 1,513[1,493 to 1,532] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Polish · high | #10 / 243Tied #1 | 1,504[1,465 to 1,544] | Unknown | Standard | 11 Nov 2026 |
| LMArena Vision (style control) LMArena · English · high | #10 / 167Tied #1 | 1,289[1,272 to 1,306] | Unknown | Standard | 9 Nov 2026 |
| LMArena Search (style control) LMArena · Overall · medium | #11 / 42 | 1,213[1,206 to 1,220] | Unknown | Standard | 9 Nov 2026 |
| LMArena Text (style control) LMArena · Industry legal and government · high | #11 / 393Tied #1 | 1,507[1,489 to 1,525] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Multi turn · high | #11 / 419 | 1,502[1,488 to 1,515] | Unknown | Standard | 11 Nov 2026 |
| LMArena Text (style control) LMArena · Industry business and management and financial operations · high | #12 / 414 | 1,493[1,481 to 1,505] | Unknown | Standard | 11 Nov 2026 |
| LMArena Vision LMArena · Creative writing vision · high | #12 / 106Tied #1 | 1,317[1,286 to 1,347] | Unknown | Standard | 9 Nov 2026 |
| Vending-Bench 2 Epoch AI Benchmarking Hub · Default | #12 / 72 | $9,235.25 | Unknown | Standard | Not stated |
| ARC-AGI-2 Epoch AI Benchmarking Hub · max | #13 / 299 | 91.7% | Unknown | Standard | Not stated |
| LMArena Search LMArena · Overall · medium | #14 / 42 | 1,213[1,207 to 1,220] | Unknown | Standard | 9 Nov 2026 |
| LMArena Text LMArena · French · high | #15 / 301 | 1,498[1,473 to 1,524] | Unknown | Standard | 11 Nov 2026 |
| GPQA diamond Epoch AI Benchmarking Hub · max · fallback | #16 / 338Tied #1 | 94.1%[91.1% to 97%] | Unknown | Standard | 22 Sept 2026 |
| Bug Hunt Bench Bug Hunt Bench · max · Claude Code | #17 / 145 | 41.7 | $58.53 | Standard | 23 Sept 2026 |
| CursorBench CursorBench · medium · Cursor | #18 / 91 | 52.5% | $2.91 | Standard | Not stated |
| LMArena Vision (style control) LMArena · Creative writing vision · high | #21 / 106 | 1,282[1,251 to 1,312] | Unknown | Standard | 9 Nov 2026 |
| Legal Agent Benchmark (HLAB-Vals) Vals AI benchmarks · max | #25 / 38 | 9.2%[4.5% to 13.8%] | $21.81 | Standard | 15 Nov 2026 |
| LMArena Text (style control) LMArena · French · high | #30 / 301 | 1,497[1,471 to 1,523] | Unknown | Standard | 11 Nov 2026 |
| ARC-AGI-2 Epoch AI Benchmarking Hub · medium | #31 / 299 | 87.5% | Unknown | Standard | Not stated |
| Bug Hunt Bench Bug Hunt Bench · xhigh · Claude Code | #32 / 145 | 36 | $34.98 | Standard | 23 Sept 2026 |
| CursorBench CursorBench · low · Cursor | #42 / 91 | 43.7% | $1.17 | Standard | Not stated |
| LMArena Vision LMArena · Humor · high | #42 / 102 | 1,267[1,226 to 1,308] | Unknown | Standard | 9 Nov 2026 |
| Bug Hunt Bench Bug Hunt Bench · high · Claude Code | #48 / 145 | 31.7 | $22.25 | Standard | 23 Sept 2026 |
| GPQA diamond Epoch AI Benchmarking Hub · max | #50 / 338 | 90.6%[86.8% to 94.4%] | Unknown | Standard | 22 Sept 2026 |
| LMArena Vision (style control) LMArena · Humor · high | #50 / 102 | 1,243[1,201 to 1,285] | Unknown | Standard | 9 Nov 2026 |
| Bug Hunt Bench Bug Hunt Bench · medium · Claude Code | #52 / 145 | 30.3 | $15.68 | Standard | 23 Sept 2026 |
| ARC-AGI-2 Epoch AI Benchmarking Hub · low | #74 / 299 | 70.1% | Unknown | Standard | Not stated |
| Bug Hunt Bench Bug Hunt Bench · low · Claude Code | #79 / 145 | 22.3 | $8.34 | Standard | 23 Sept 2026 |
Same benchmark and harness measured across multiple reasoning effort settings.
Epoch AI Benchmarking Hub
70.1
87.5
93.3
92.5
91.7
Bug Hunt Bench · Claude Code
22.3
$8.34
30.3
$15.68
31.7
$22.25
36.0
$34.98
41.7
$58.53
CursorBench · Cursor
43.7
$1.17
52.5
$2.91
56.0
$3.97
56.0
$6.98
57.8
$13.43
LMArena
0.1
$2.12
0.2
$5.65
Direct pairwise win rates for Opus 5.5 across shared benchmark boards.
22 shared boards
73% win
Head-to-head22 shared boards
73% win
Head-to-head21 shared boards
100% win
Head-to-head21 shared boards
90% win
Head-to-head21 shared boards
62% win
Head-to-head20 shared boards
95% win
Head-to-head20 shared boards
90% win
Head-to-head19 shared boards
58% win
Head-to-head19 shared boards
95% win
Head-to-head18 shared boards
94% win
Head-to-head