LLM Debate Benchmark
Models debate an opponent over multiple rounds. Judges compare the arguments.
The author publishes the model comparison in the repository.
- Made by
- Lech Mazur
- Kind
- Multi-turn conversation
- Level
- University
- Scored
- Ranked by matches
- Answers
- Open-ended
- Tasks
- All public
- Measures
- Preference
- Language
- English
- Interaction
- Multi-agent
- Items from
- Live play or votes
- Refresh
- Live
- Run by
- Anyone
- In views
- Agents, Kept fresh
- Strongest model tested
- GPT-6.2 Astra #1 on the capability index
- Last model added
- November 2026
- Scores refreshed
- November 20th, 2026
Claude Fable 5.1 leads with 1,706 Claude Fable 5.1 leads with 1,706
Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. This view shows only models we track, each at its best.
- 1,706 Claude Fable 5.1
- 1,705 Claude Fable 5
- 1,704 Claude Opus 5
- 1,688 Claude Fable 5.2
- 1,680 Claude Opus 5.5
- 1,679
Qwen4-Max
- 1,679 Kimi K3
- 1,676
Gemini 4 Carbon
- 1,673 GPT-6.2 Astra
- 1,649
Gemini 4 Argon
- 1,647
Tencent Hy4
- 1,637
Muse Wonder
- 1,627 Claude Opus 4.7
- 1,625 GPT-5.6 Sol
- 1,617
Muse Spark 1.1
- 1,613 GLM-5.3
- 1,610 Claude Opus 4.8
- 1,607 GPT-6 Astra
- 1,601 Grok 4.7
- 1,601 Kimi K3.1 Pro
- 1,585 Grok 4.6
- 1,584 Xiaomi MiMo V2.6 Pro
- 1,579
DeepSeek V4.1 Flash 1017
- 1,578
Muse Spark 1.3
- 1,568 Claude Sonnet 5
- 1,564
DeepSeek V4.1 Flash
- 1,561
Tencent Hy4 preview
- 1,552
DeepSeek V4 Pro 0813
- 1,542 GLM-5.2
- 1,540 Claude Sonnet 4.6
- 1,535
Qwen 3.8 Max
- 1,525 GPT-5.4
- 1,521
Step 5 Preview
- 1,509
MiniMax-M3.1-Flash
- 1,507 GPT-5.5
- 1,498 GLM-5.1
- 1,490 Kimi K2.6
- 1,488
Gemini 3.8 Flash
- 1,483
MiniMax-M3
- 1,470 Grok 4.5
- 1,468
Gemini 3.1 Pro
- 1,468 ByteDance Seed2.1 Pro
- 1,463 Xiaomi MiMo V2.5 Pro
- 1,454
Qwen 3.6 Max (Preview)
- 1,442
Mistral Large 4
- 1,434
Gemini 3.7 Flash
- 1,434 ByteDance Seed2.0 Pro
- 1,431 Kimi K2.5
- 1,427
DeepSeek-V4-Pro
- 1,414
Qwen3.7-Max
- 1,412
MiniMax-M2.7
- 1,390 Grok 4.20 0309
- 1,387
Gemini 3.5 Flash
- 1,370 Baidu Ernie 5.1
- 1,369 Xiaomi MiMo V2 Pro
- 1,368
Qwen3.5 397B-A17B
- 1,366
Tencent Hy3 preview
- 1,355
Step 3.7 Flash
- 1,350 Grok 4.3
- 1,341
DeepSeek-V3.2
- 1,321
Mistral Medium 3.5
- 1,308
Gemini 3.1 Flash-Lite
- 1,242 gpt-oss-120b
- 1,225 Baidu Ernie 5.0
- 1,170
Mistral Large 3
- 1,006
Llama 4 Maverick
Bars start at 1,000.
- 1 Claude Fable 5.1 high 1,706
- 2 Claude Fable 5 high 1,705
- 3 Claude Opus 5 high 1,704
- 4 Claude Fable 5.2 high 1,688
- 5 Claude Opus 5.5 high 1,680
- 6
Qwen4-Max high 1,679
- 7 Kimi K3 1,679
- 8
Gemini 4 Carbon high 1,676
- 9 GPT-6.2 Astra high 1,673
- 10
Gemini 4 Argon high 1,649
- 11
Tencent Hy4 high 1,647
- 12
Muse Wonder high 1,637
- 13 Claude Opus 4.7 high 1,627
- 14 GPT-5.6 Sol high 1,625
- 15
Muse Spark 1.1 high 1,617
- 16 GLM-5.3 high 1,613
- 17 Claude Opus 4.8 high 1,610
- 18 GPT-6 Astra high 1,607
- 19 Grok 4.7 high 1,601
- 20 Kimi K3.1 Pro high 1,601
- 21 Grok 4.6 high 1,585
- 22 Xiaomi MiMo V2.6 Pro thinking 1,584
- 23
DeepSeek V4.1 Flash 1017 high 1,579
- 24
Muse Spark 1.3 high 1,578
- 25 Claude Sonnet 5 high 1,568
- 26
DeepSeek V4.1 Flash high 1,564
- 27
Tencent Hy4 preview high 1,561
- 28
DeepSeek V4 Pro 0813 high 1,552
- 29 GLM-5.2 max 1,542
- 30 Claude Sonnet 4.6 high 1,540
- 31
Qwen 3.8 Max 1,535
- 32 GPT-5.4 high 1,525
- 33
Step 5 Preview high 1,521
- 34
MiniMax-M3.1-Flash high 1,509
- 35 GPT-5.5 high 1,507
- 36 GLM-5.1 1,498
- 37 Kimi K2.6 1,490
- 38
Gemini 3.8 Flash high 1,488
- 39
MiniMax-M3 1,483
- 40 Grok 4.5 high 1,470
- 41
Gemini 3.1 Pro 1,468
- 42 ByteDance Seed2.1 Pro 1,468
- 43 Xiaomi MiMo V2.5 Pro 1,463
- 44
Qwen 3.6 Max (Preview) 1,454
- 45
Mistral Large 4 high 1,442
- 46
Gemini 3.7 Flash high 1,434
- 47 ByteDance Seed2.0 Pro 1,434
- 48 Kimi K2.5 thinking 1,431
- 49
DeepSeek-V4-Pro 1,427
- 50
Qwen3.7-Max 1,414
- 51
MiniMax-M2.7 1,412
- 52 Grok 4.20 0309 Reasoning 1,390
- 53
Gemini 3.5 Flash 1,387
- 54 Baidu Ernie 5.1 1,370
- 55 Xiaomi MiMo V2 Pro 1,369
- 56
Qwen3.5 397B-A17B 1,368
- 57
Tencent Hy3 preview high 1,366
- 58
Step 3.7 Flash high 1,355
- 59 Grok 4.3 1,350
- 60
DeepSeek-V3.2 1,341
- 61
Mistral Medium 3.5 high 1,321
- 62
Gemini 3.1 Flash-Lite 1,308
- 63 gpt-oss-120b 1,242
- 64 Baidu Ernie 5.0 1,225
- 65
Mistral Large 3 1,170
- 66
Llama 4 Maverick 1,006
Source: Official LLM Debate Benchmark leaderboard
- 1,706 Claude Fable 5.1
- 1,705 Claude Fable 5
- 1,704 Claude Opus 5
- 1,688 Claude Fable 5.2
- 1,680 Claude Opus 5.5
- 1,679
Qwen4-Max
- 1,679 Kimi K3
- 1,676
Gemini 4 Carbon
- 1,673 GPT-6.2 Astra
- 1,649
Gemini 4 Argon
- 1,647
Tencent Hy4
- 1,637
Muse Wonder
- 1,627 Claude Opus 4.7
- 1,625 GPT-5.6 Sol
- 1,617
Muse Spark 1.1
- 1,613 GLM-5.3
- 1,610 Claude Opus 4.8
- 1,607 GPT-6 Astra
- 1,601 Grok 4.7
- 1,601 Kimi K3.1 Pro
- 1,585 Grok 4.6
- 1,579
DeepSeek V4.1 Flash 1017
- 1,578
Muse Spark 1.3
- 1,568 Claude Sonnet 5
- 1,564
DeepSeek V4.1 Flash
- 1,561
Tencent Hy4 preview
- 1,552
DeepSeek V4 Pro 0813
- 1,542 GLM-5.2
- 1,540 Claude Sonnet 4.6
- 1,535
Qwen 3.8 Max
- 1,525 GPT-5.4
- 1,521
Step 5 Preview
- 1,509
MiniMax-M3.1-Flash
- 1,507 GPT-5.5
- 1,498 GLM-5.1
- 1,490 Kimi K2.6
- 1,488
Gemini 3.8 Flash
- 1,483
MiniMax-M3
- 1,470 Grok 4.5
- 1,468
Gemini 3.1 Pro
- 1,454
Qwen 3.6 Max (Preview)
- 1,442
Mistral Large 4
- 1,434
Gemini 3.7 Flash
- 1,431 Kimi K2.5
- 1,427
DeepSeek-V4-Pro
- 1,414
Qwen3.7-Max
- 1,412
MiniMax-M2.7
- 1,387
Gemini 3.5 Flash
- 1,368
Qwen3.5 397B-A17B
- 1,366
Tencent Hy3 preview
- 1,355
Step 3.7 Flash
- 1,350 Grok 4.3
- 1,341
DeepSeek-V3.2
- 1,321
Mistral Medium 3.5
- 1,308
Gemini 3.1 Flash-Lite
- 1,242 gpt-oss-120b
- 1,170
Mistral Large 3
- 1,006
Llama 4 Maverick
Bars start at 1,000.
- 1 Claude Fable 5.1 high 1,706
- 2 Claude Fable 5 high 1,705
- 3 Claude Opus 5 high 1,704
- 4 Claude Fable 5.2 high 1,688
- 5 Claude Opus 5.5 high 1,680
- 6
Qwen4-Max high 1,679
- 7 Kimi K3 1,679
- 8
Gemini 4 Carbon high 1,676
- 9 GPT-6.2 Astra high 1,673
- 10
Gemini 4 Argon high 1,649
- 11
Tencent Hy4 high 1,647
- 12
Muse Wonder high 1,637
- 13 Claude Opus 4.7 high 1,627
- 14 GPT-5.6 Sol high 1,625
- 15
Muse Spark 1.1 high 1,617
- 16 GLM-5.3 high 1,613
- 17 Claude Opus 4.8 high 1,610
- 18 GPT-6 Astra high 1,607
- 19 Grok 4.7 high 1,601
- 20 Kimi K3.1 Pro high 1,601
- 21 Grok 4.6 high 1,585
- 22
DeepSeek V4.1 Flash 1017 high 1,579
- 23
Muse Spark 1.3 high 1,578
- 24 Claude Sonnet 5 high 1,568
- 25
DeepSeek V4.1 Flash high 1,564
- 26
Tencent Hy4 preview high 1,561
- 27
DeepSeek V4 Pro 0813 high 1,552
- 28 GLM-5.2 max 1,542
- 29 Claude Sonnet 4.6 high 1,540
- 30
Qwen 3.8 Max 1,535
- 31 GPT-5.4 high 1,525
- 32
Step 5 Preview high 1,521
- 33
MiniMax-M3.1-Flash high 1,509
- 34 GPT-5.5 high 1,507
- 35 GLM-5.1 1,498
- 36 Kimi K2.6 1,490
- 37
Gemini 3.8 Flash high 1,488
- 38
MiniMax-M3 1,483
- 39 Grok 4.5 high 1,470
- 40
Gemini 3.1 Pro 1,468
- 41
Qwen 3.6 Max (Preview) 1,454
- 42
Mistral Large 4 high 1,442
- 43
Gemini 3.7 Flash high 1,434
- 44 Kimi K2.5 thinking 1,431
- 45
DeepSeek-V4-Pro 1,427
- 46
Qwen3.7-Max 1,414
- 47
MiniMax-M2.7 1,412
- 48
Gemini 3.5 Flash 1,387
- 49
Qwen3.5 397B-A17B 1,368
- 50
Tencent Hy3 preview high 1,366
- 51
Step 3.7 Flash high 1,355
- 52 Grok 4.3 1,350
- 53
DeepSeek-V3.2 1,341
- 54
Mistral Medium 3.5 high 1,321
- 55
Gemini 3.1 Flash-Lite 1,308
- 56 gpt-oss-120b 1,242
- 57
Mistral Large 3 1,170
- 58
Llama 4 Maverick 1,006
Source: Official LLM Debate Benchmark leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude Fable 5.1 | 1,706 | high | Sep 2026 | Official leaderboard |
| 2 | Claude Fable 5 | 1,705 | high | Jun 2026 | Official leaderboard |
| 3 | Claude Opus 5 | 1,704 | high | Jul 2026 | Official leaderboard |
| 4 | Claude Fable 5.2 | 1,688 | high | Nov 2026 | Official leaderboard |
| 5 | Claude Opus 5.5 | 1,680 | high | Sep 2026 | Official leaderboard |
| 6 | | 1,679 | high | Nov 2026 | Official leaderboard |
| 7 | Kimi K3 | 1,679 | Jul 2026 | Official leaderboard | |
| 8 | | 1,676 | high | Oct 2026 | Official leaderboard |
| 9 | GPT-6.2 Astra | 1,673 | high | Nov 2026 | Official leaderboard |
| 10 | | 1,649 | high | Sep 2026 | Official leaderboard |
| 11 | | 1,647 | high | Nov 2026 | Official leaderboard |
| 12 | | 1,637 | high | Oct 2026 | Official leaderboard |
| 13 | Claude Opus 4.7 | 1,627 | high | Apr 2026 | Official leaderboard |
| 14 | GPT-5.6 Sol | 1,625 | high | Jul 2026 | Official leaderboard |
| 15 | | 1,617 | high | Jul 2026 | Official leaderboard |
| 16 | GLM-5.3 | 1,613 | high | Aug 2026 | Official leaderboard |
| 17 | Claude Opus 4.8 | 1,610 | high | May 2026 | Official leaderboard |
| 18 | GPT-6 Astra | 1,607 | high | Sep 2026 | Official leaderboard |
| 19 | Grok 4.7 | 1,601 | high | Sep 2026 | Official leaderboard |
| 20 | Kimi K3.1 Pro | 1,601 | high | Oct 2026 | Official leaderboard |
| 21 | Grok 4.6 | 1,585 | high | Aug 2026 | Official leaderboard |
| 22 | Xiaomi MiMo V2.6 Pro | 1,584 | thinking | Official leaderboard | |
| 23 | | 1,579 | high | Oct 2026 | Official leaderboard |
| 24 | | 1,578 | high | Sep 2026 | Official leaderboard |
| 25 | Claude Sonnet 5 | 1,568 | high | Jun 2026 | Official leaderboard |
| 26 | | 1,564 | high | Sep 2026 | Official leaderboard |
| 27 | | 1,561 | high | Aug 2026 | Official leaderboard |
| 28 | | 1,552 | high | Aug 2026 | Official leaderboard |
| 29 | GLM-5.2 | 1,542 | max | Jun 2026 | Official leaderboard |
| 30 | Claude Sonnet 4.6 | 1,540 | high | Feb 2026 | Official leaderboard |
| 31 | | 1,535 | Aug 2026 | Official leaderboard | |
| 32 | GPT-5.4 | 1,525 | high | Mar 2026 | Official leaderboard |
| 33 | | 1,521 | high | Sep 2026 | Official leaderboard |
| 34 | | 1,509 | high | Oct 2026 | Official leaderboard |
| 35 | GPT-5.5 | 1,507 | high | Apr 2026 | Official leaderboard |
| 36 | GLM-5.1 | 1,498 | Apr 2026 | Official leaderboard | |
| 37 | Kimi K2.6 | 1,490 | Apr 2026 | Official leaderboard | |
| 38 | | 1,488 | high | Sep 2026 | Official leaderboard |
| 39 | | 1,483 | Jun 2026 | Official leaderboard | |
| 40 | Grok 4.5 | 1,470 | high | Jul 2026 | Official leaderboard |
| 41 | | 1,468 | Feb 2026 | Official leaderboard | |
| 42 | ByteDance Seed2.1 Pro | 1,468 | Official leaderboard | ||
| 43 | Xiaomi MiMo V2.5 Pro | 1,463 | Official leaderboard | ||
| 44 | | 1,454 | Apr 2026 | Official leaderboard | |
| 45 | | 1,442 | high | Oct 2026 | Official leaderboard |
| 46 | | 1,434 | high | Aug 2026 | Official leaderboard |
| 47 | ByteDance Seed2.0 Pro | 1,434 | Official leaderboard | ||
| 48 | Kimi K2.5 | 1,431 | thinking | Jan 2026 | Official leaderboard |
| 49 | | 1,427 | Apr 2026 | Official leaderboard | |
| 50 | | 1,414 | May 2026 | Official leaderboard | |
| 51 | | 1,412 | Mar 2026 | Official leaderboard | |
| 52 | Grok 4.20 0309 | 1,390 | Reasoning | Official leaderboard | |
| 53 | | 1,387 | May 2026 | Official leaderboard | |
| 54 | Baidu Ernie 5.1 | 1,370 | Official leaderboard | ||
| 55 | Xiaomi MiMo V2 Pro | 1,369 | Official leaderboard | ||
| 56 | | 1,368 | Feb 2026 | Official leaderboard | |
| 57 | | 1,366 | high | Apr 2026 | Official leaderboard |
| 58 | | 1,355 | high | May 2026 | Official leaderboard |
| 59 | Grok 4.3 | 1,350 | Apr 2026 | Official leaderboard | |
| 60 | | 1,341 | Dec 2025 | Official leaderboard | |
| 61 | | 1,321 | high | Apr 2026 | Official leaderboard |
| 62 | | 1,308 | Mar 2026 | Official leaderboard | |
| 63 | gpt-oss-120b | 1,242 | Aug 2025 | Official leaderboard | |
| 64 | Baidu Ernie 5.0 | 1,225 | Official leaderboard | ||
| 65 | | 1,170 | Dec 2025 | Official leaderboard | |
| 66 | | 1,006 | Apr 2025 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude Fable 5.1 | 1,706 | high | Sep 2026 | Official leaderboard |
| 2 | Claude Fable 5 | 1,705 | high | Jun 2026 | Official leaderboard |
| 3 | Claude Opus 5 | 1,704 | high | Jul 2026 | Official leaderboard |
| 4 | Claude Fable 5.2 | 1,688 | high | Nov 2026 | Official leaderboard |
| 5 | Claude Opus 5.5 | 1,680 | high | Sep 2026 | Official leaderboard |
| 6 | | 1,679 | high | Nov 2026 | Official leaderboard |
| 7 | Kimi K3 | 1,679 | Jul 2026 | Official leaderboard | |
| 8 | | 1,676 | high | Oct 2026 | Official leaderboard |
| 9 | GPT-6.2 Astra | 1,673 | high | Nov 2026 | Official leaderboard |
| 10 | | 1,649 | high | Sep 2026 | Official leaderboard |
| 11 | | 1,647 | high | Nov 2026 | Official leaderboard |
| 12 | | 1,637 | high | Oct 2026 | Official leaderboard |
| 13 | Claude Opus 4.7 | 1,627 | high | Apr 2026 | Official leaderboard |
| 14 | GPT-5.6 Sol | 1,625 | high | Jul 2026 | Official leaderboard |
| 15 | | 1,617 | high | Jul 2026 | Official leaderboard |
| 16 | GLM-5.3 | 1,613 | high | Aug 2026 | Official leaderboard |
| 17 | Claude Opus 4.8 | 1,610 | high | May 2026 | Official leaderboard |
| 18 | GPT-6 Astra | 1,607 | high | Sep 2026 | Official leaderboard |
| 19 | Grok 4.7 | 1,601 | high | Sep 2026 | Official leaderboard |
| 20 | Kimi K3.1 Pro | 1,601 | high | Oct 2026 | Official leaderboard |
| 21 | Grok 4.6 | 1,585 | high | Aug 2026 | Official leaderboard |
| 22 | | 1,579 | high | Oct 2026 | Official leaderboard |
| 23 | | 1,578 | high | Sep 2026 | Official leaderboard |
| 24 | Claude Sonnet 5 | 1,568 | high | Jun 2026 | Official leaderboard |
| 25 | | 1,564 | high | Sep 2026 | Official leaderboard |
| 26 | | 1,561 | high | Aug 2026 | Official leaderboard |
| 27 | | 1,552 | high | Aug 2026 | Official leaderboard |
| 28 | GLM-5.2 | 1,542 | max | Jun 2026 | Official leaderboard |
| 29 | Claude Sonnet 4.6 | 1,540 | high | Feb 2026 | Official leaderboard |
| 30 | | 1,535 | Aug 2026 | Official leaderboard | |
| 31 | GPT-5.4 | 1,525 | high | Mar 2026 | Official leaderboard |
| 32 | | 1,521 | high | Sep 2026 | Official leaderboard |
| 33 | | 1,509 | high | Oct 2026 | Official leaderboard |
| 34 | GPT-5.5 | 1,507 | high | Apr 2026 | Official leaderboard |
| 35 | GLM-5.1 | 1,498 | Apr 2026 | Official leaderboard | |
| 36 | Kimi K2.6 | 1,490 | Apr 2026 | Official leaderboard | |
| 37 | | 1,488 | high | Sep 2026 | Official leaderboard |
| 38 | | 1,483 | Jun 2026 | Official leaderboard | |
| 39 | Grok 4.5 | 1,470 | high | Jul 2026 | Official leaderboard |
| 40 | | 1,468 | Feb 2026 | Official leaderboard | |
| 41 | | 1,454 | Apr 2026 | Official leaderboard | |
| 42 | | 1,442 | high | Oct 2026 | Official leaderboard |
| 43 | | 1,434 | high | Aug 2026 | Official leaderboard |
| 44 | Kimi K2.5 | 1,431 | thinking | Jan 2026 | Official leaderboard |
| 45 | | 1,427 | Apr 2026 | Official leaderboard | |
| 46 | | 1,414 | May 2026 | Official leaderboard | |
| 47 | | 1,412 | Mar 2026 | Official leaderboard | |
| 48 | | 1,387 | May 2026 | Official leaderboard | |
| 49 | | 1,368 | Feb 2026 | Official leaderboard | |
| 50 | | 1,366 | high | Apr 2026 | Official leaderboard |
| 51 | | 1,355 | high | May 2026 | Official leaderboard |
| 52 | Grok 4.3 | 1,350 | Apr 2026 | Official leaderboard | |
| 53 | | 1,341 | Dec 2025 | Official leaderboard | |
| 54 | | 1,321 | high | Apr 2026 | Official leaderboard |
| 55 | | 1,308 | Mar 2026 | Official leaderboard | |
| 56 | gpt-oss-120b | 1,242 | Aug 2025 | Official leaderboard | |
| 57 | | 1,170 | Dec 2025 | Official leaderboard | |
| 58 | | 1,006 | Apr 2025 | Official leaderboard |
The best score, 1,706, costs $0.39 a task
Runs on the cost frontier score higher than any run that's cheaper. Qwen4-Max at high comes within 26.7 of the best for $0.07 a task, 5 times cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.
Source: Official LLM Debate Benchmark leaderboard
The best score, 1,706, costs $0.39 a task
Runs on the cost frontier score higher than any run that's cheaper. Qwen4-Max at high comes within 26.7 of the best for $0.07 a task, 5 times cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.
Source: Official LLM Debate Benchmark leaderboard
8 models have held number one on the official leaderboard since 2023
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official LLM Debate Benchmark leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official LLM Debate Benchmark leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Instructions and chat
- IFEval In use Models answer about 500 prompts containing verifiable instructions such as word limits or required keywords. Code checks each instruction for strict and loose accuracy.
Gemma 4 31B IT 98.9%
- IFBench In use 58 new out-of-domain verifiable output constraints. Scored by programmatic constraint checks (strict or loose). Flint 89.2%
- MultiChallenge In use Multi-turn conversations testing instruction retention, inference memory, versioned editing and self-coherence. Graded by LLM judge with rubrics. GPT-6.1 Sol 82%
- Multi-IF In use Models follow verifiable instructions over three-turn conversations in eight languages, extending IFEval. Scored by instruction accuracy at each turn. LFM3-1.7B 89.1%
- AdvancedIF In use Models respond to over 1,600 prompts with complex, multi-turn, and system-level instructions. A judge model checks each response against expert-written rubrics.
MAI-Thinking-1 85%
- MM-IFEval In use Multimodal models follow image-grounded instructions with exact output constraints. Checks mix rule-based verification and a judge model. GPT-4o 64.6%
Data audit
All 10 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 4 pass 1 note
- Pass Independent runs Scores from the official leaderboard (66 entries).
- Pass Models scored 58 models we track, plus 8 other entries on its leaderboard.
- Pass Newest score The newest is from November 2026, for GPT-6.2 Astra.
- Pass Frontier models 7 of the top 10 have a score, led by GPT-6.2 Astra (#1).
- Note Headroom Scored as a rating, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on github.com: 60 rows for models we track and 8 other entries at the last read.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.