Benchmark Almanac

MultiChallenge

GPT-6.1 Sol 82% leads since September 2026
GPT-6.1 Sol 82% leads since September 2026
In use

Multi-turn conversations testing instruction retention, inference memory, versioned editing and self-coherence. Graded by LLM judge with rubrics.

On February 3rd, 2026 Scale switched the judge to Gemini 2.5 Pro and rewrote the rubric questions of 54 tasks. Scores published before that date may not match the current board.

Made by
Scale AI
Released
January 2025
Announced on X
16K views, 26 likes
Kind
Multi-turn conversation
Level
Everyday
Scored
Graded by a model
Answers
Open-ended
Tasks
Public, plus a private test set
Measures
Capability
Language
English
In and out
Text → Text
Interaction
Multi-turn
Refresh
One fixed set
Run by
A host
Strongest model tested
GPT-6.2 Astra #1 on the capability index
Last in a launch post
October 2026
Last model added
November 2026
Scores refreshed
November 19th, 2026

In the authors’ words

We present MultiChallenge, a pioneering benchmark evaluating large language models (LLMs) on conducting multi-turn conversations with human users, a crucial yet underexamined capability for their applications. MultiChallenge identifies four categories of challenges in multi-turn conversations that are not only common and realistic among current human-LLM interactions, but are also challenging to all current frontier LLMs. All 4 challenges require accurate instruction-following, context allocation, and in-context reasoning at the same time. We also develop LLM as judge with instance-level rubrics to facilitate an automatic evaluation method with fair agreement with experienced human raters. Despite achieving near-perfect scores on existing multi-turn evaluation benchmarks, all frontier models have less than 50% accuracy on MultiChallenge, with the top-performing Claude 3.5 Sonnet (June 2024) achieving just a 41.4% average accuracy.

Abstract of MultiChallenge: A Realistic Multi-Turn Conversation Evaluation Benchmark Challenging to Frontier LLMs (arXiv:2501.17399, CC0)
47
Entries on its own leaderboard, 45 models we track
36
Launch posts list it, from 13 labs
21
System cards report it, from 11 labs

GPT-6.1 Sol leads with 82% GPT-6.1 Sol leads with 82%

Best score per entry from the benchmark's own leaderboard, read on November 19th, 2026. Where a model ran with several settings or harnesses, we keep its best. 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.

GPT-6.1 Sol leads with 82%, claims included

Best score per model from every source, including labs’ own numbers and claims Epoch AI collected from reports and papers. Labs pick their own settings, so a claim can beat an independent run of the same model. Striped bars are claims, and dotted bars are reruns by Epoch AI or another board.

  1. 1 GPT-6.1 Sol 82%
  2. 2 Gemini 4 Hydrogen 80.6%
  3. 3 Gemini 4 Argon 78.7%
  4. 4 Muse Spark 75.5%
  5. 5 Muse Spark 1.1 75.3%
  6. 6 DeepSeek V4.1 Flash 1017 74.9%
  7. 7 Muse Spark Open 74.2%
  8. 8 GPT-6.1 Terra 74.2%
  9. 9 GPT-6.2 Astra 73.2%
  10. 10 GPT-6.2 Sol 72%
  11. 11 Gemini 3.1 Pro 71.4%
  12. 12 GPT-5.4 Pro 69.2%
  13. 13 Claude Haiku 5.5 67.7%
  14. 14 Gemini 3 Pro 65.7%
  15. 15 GPT-5.1 thinking 63.4%
  16. 16 GPT-5 thinking 63.2%
  17. 17 o3-pro high 62.4%
  18. 18 Kimi K2.5 61.4%
  19. 19 Gemini 3.1 Flash-Lite 60.6%
  20. 20 GPT-5 mini thinking 59%
  21. 21 Claude Opus 4.5 thinking 59%
  22. 22 Claude Opus 4 thinking 58.6%
  23. 23 Claude Opus 4.1 thinking 57.2%
  24. 24 Claude Sonnet 4 thinking 57.1%
  25. 25 o3 high 56.6%
  26. 26 Nemotron 3.5 Bolt E4B thinking 56%
  27. 27 Claude Opus 4.6 non-thinking 56%
  28. 28 Kimi K2 Thinking 55.4%
  29. 29 Claude Sonnet 4.5 thinking 55.3%
  30. 30 Gemini 2.5 Pro Jun 2025 53.6%
  31. 31 Claude 3.7 Sonnet thinking 51.6%
  32. 32 GPT-5.1 Instant 51.2%
  33. 33 Claude Haiku 4.5 thinking 50.5%
  34. 34 DeepSeek-V3.1 46.1%
  35. 35 gpt-oss-120b 45.3%
  36. 36 o4-mini high 44.9%
  37. 37 Gemma 4 E4B thinking 43.2%
  38. 38 Qwen3-235B-A22B 41.2%
  39. 39 GPT-4.1 39.4%
  40. 40 Nemotron 3 Nano 4B thinking 38.7%
  41. 41 Gemini 2.0 Flash Feb 2025 36.4%
  42. 42 Gemma 4 E2B thinking 36%
  43. 43 Ministral 3 8B thinking 27.8%
  44. 44 Ministral 3 3B thinking 25.7%
  45. 45 Qwen3.5-4B thinking 25.1%
  46. 46 Llama 3.1-8B 23.7%
  47. 47 Gemma 3 4B 22.8%

Source: Official MultiChallenge leaderboard

Source: Official MultiChallenge leaderboard

80 models. Scroll sideways for the rest.

  1. 82.0 GPT-6.1 Sol
  2. 80.6 Gemini 4 Hydrogen
  3. 78.7 Gemini 4 Argon
  4. 77.7 Nova 2.0 Pro Preview
  5. 76.6 Nova 2 Lite
  6. 75.5 Muse Spark
  7. 75.5 Nova 2 Omni Preview
  8. 75.3 Muse Spark 1.1
  9. 74.9 DeepSeek V4.1 Flash 1017
  10. 74.2 Muse Spark Open
  11. 74.2 GPT-6.1 Terra
  12. 73.2 GPT-6.2 Astra
  13. 72.0 GPT-6.2 Sol
  14. 71.4 Gemini 3.1 Pro
  15. 69.9 Seed2.0 Lite (0428)
  16. 69.8 Ling-3.1-flash
  17. 69.6 GPT-5
  18. 69.2 GPT-5.4 Pro
  19. 68.3 Seed2.0 Pro
  20. 67.7 Claude Haiku 5.5
  21. 67.6 Qwen3.5 397B-A17B
  22. 66.7 Seed1.8
  23. 66.0 ERNIE 5.0
  24. 65.7 Gemini 3 Pro
  25. 64.8 GLM-5
  26. 63.8 Nemotron 3 Ultra
  27. 63.4 GPT-5.1
  28. 63.3 Qwen3-Max-2026-01-23
  29. 62.6 STEP3-VL-10B
  30. 62.6 Tencent HY 2.0 Think
  31. 62.4 o3-pro
  32. 62.3 GPT-5 mini
  33. 61.5 Qwen3.5-122B-A10B
  34. 61.4 Kimi K2.5
  35. 61.0 Solar Open 2
  36. 60.8 Qwen3.5-27B
  37. 60.6 Gemini 3.1 Flash-Lite
  38. 60.4 Mercury 3
  39. 60.0 Qwen3.5-35B-A3B
  40. 59.0 Claude Opus 4.5
  41. 58.6 Claude Opus 4
  42. 57.2 Claude Opus 4.1
  43. 57.1 Claude Sonnet 4
  44. 56.6 o3
  45. 56.0 Nemotron 3.5 Bolt E4B
  46. 56.0 Claude Opus 4.6
  47. 55.7 Step 3.5 Flash
  48. 55.4 Kimi K2 Thinking
  49. 55.3 Claude Sonnet 4.5
  50. 55.2 Nemotron 3 Super
  51. 54.9 GPT-5 nano
  52. 54.5 Qwen3.5-9B
  53. 54.1 Kimi K2
  54. 53.6 Gemini 2.5 Pro
  55. 53.0 MAI-Thinking-1
  56. 52.8 GLM-4.5
  57. 51.6 Claude 3.7 Sonnet
  58. 51.2 GPT-5.1 Instant
  59. 50.9 Ring-1T
  60. 50.5 Claude Haiku 4.5
  61. 49.0 Qwen3.5-4B
  62. 46.1 DeepSeek-V3.1
  63. 45.3 gpt-oss-120b
  64. 45.3 Nemotron-Cascade 2 30B-A3B
  65. 44.9 o4-mini
  66. 44.7 MiniMax-M1-80k
  67. 43.2 Gemma 4 E4B
  68. 42.5 GLM-4.5-Air
  69. 41.2 Qwen3-235B-A22B
  70. 39.4 GPT-4.1
  71. 38.7 Nemotron 3 Nano 4B
  72. 38.5 Nemotron 3 Nano 30B-A3B
  73. 36.4 Gemini 2.0 Flash
  74. 36.0 Gemma 4 E2B
  75. 35.8 GPT-4.1 mini
  76. 33.7 Qwen3.5-2B
  77. 23.7 Llama 3.1-8B
  78. 22.8 Gemma 3 4B
  79. 18.9 Qwen3.5-0.8B
  80. 15.0 GPT-4.1 nano
  1. 1 GPT-6.1 Sol 82%
  2. 2 Gemini 4 Hydrogen 80.6%
  3. 3 Gemini 4 Argon 78.7%
  4. 4 Nova 2.0 Pro Preview 77.7%
  5. 5 Nova 2 Lite 76.6%
  6. 6 Muse Spark 75.5%
  7. 7 Nova 2 Omni Preview 75.5%
  8. 8 Muse Spark 1.1 75.3%
  9. 9 DeepSeek V4.1 Flash 1017 74.9%
  10. 10 Muse Spark Open 74.2%
  11. 11 GPT-6.1 Terra 74.2%
  12. 12 GPT-6.2 Astra 73.2%
  13. 13 GPT-6.2 Sol 72%
  14. 14 Gemini 3.1 Pro 71.4%
  15. 15 Seed2.0 Lite (0428) 69.9%
  16. 16 Ling-3.1-flash 69.8%
  17. 17 GPT-5 69.6%
  18. 18 GPT-5.4 Pro 69.2%
  19. 19 Seed2.0 Pro 68.3%
  20. 20 Claude Haiku 5.5 67.7%
  21. 21 Qwen3.5 397B-A17B 67.6%
  22. 22 Seed1.8 66.7%
  23. 23 ERNIE 5.0 66%
  24. 24 Gemini 3 Pro 65.7%
  25. 25 GLM-5 64.8%
  26. 26 Nemotron 3 Ultra 63.8%
  27. 27 GPT-5.1 thinking 63.4%
  28. 28 Qwen3-Max-2026-01-23 63.3%
  29. 29 STEP3-VL-10B 62.6%
  30. 30 Tencent HY 2.0 Think 62.6%
  31. 31 o3-pro high 62.4%
  32. 32 GPT-5 mini 62.3%
  33. 33 Qwen3.5-122B-A10B 61.5%
  34. 34 Kimi K2.5 61.4%
  35. 35 Solar Open 2 61%
  36. 36 Qwen3.5-27B 60.8%
  37. 37 Gemini 3.1 Flash-Lite 60.6%
  38. 38 Mercury 3 60.4%
  39. 39 Qwen3.5-35B-A3B 60%
  40. 40 Claude Opus 4.5 thinking 59%
  41. 41 Claude Opus 4 thinking 58.6%
  42. 42 Claude Opus 4.1 thinking 57.2%
  43. 43 Claude Sonnet 4 thinking 57.1%
  44. 44 o3 high 56.6%
  45. 45 Nemotron 3.5 Bolt E4B thinking 56%
  46. 46 Claude Opus 4.6 non-thinking 56%
  47. 47 Step 3.5 Flash 55.7%
  48. 48 Kimi K2 Thinking 55.4%
  49. 49 Claude Sonnet 4.5 thinking 55.3%
  50. 50 Nemotron 3 Super 55.2%
  51. 51 GPT-5 nano 54.9%
  52. 52 Qwen3.5-9B 54.5%
  53. 53 Kimi K2 Jul 2025 54.1%
  54. 54 Gemini 2.5 Pro Jun 2025 53.6%
  55. 55 MAI-Thinking-1 53%
  56. 56 GLM-4.5 52.8%
  57. 57 Claude 3.7 Sonnet thinking 51.6%
  58. 58 GPT-5.1 Instant 51.2%
  59. 59 Ring-1T 50.9%
  60. 60 Claude Haiku 4.5 thinking 50.5%
  61. 61 Qwen3.5-4B 49%
  62. 62 DeepSeek-V3.1 46.1%
  63. 63 gpt-oss-120b 45.3%
  64. 64 Nemotron-Cascade 2 30B-A3B 45.3%
  65. 65 o4-mini high 44.9%
  66. 66 MiniMax-M1-80k 44.7%
  67. 67 Gemma 4 E4B thinking 43.2%
  68. 68 GLM-4.5-Air 42.5%
  69. 69 Qwen3-235B-A22B 41.2%
  70. 70 GPT-4.1 39.4%
  71. 71 Nemotron 3 Nano 4B thinking 38.7%
  72. 72 Nemotron 3 Nano 30B-A3B 38.5%
  73. 73 Gemini 2.0 Flash Feb 2025 36.4%
  74. 74 Gemma 4 E2B thinking 36%
  75. 75 GPT-4.1 mini 35.8%
  76. 76 Qwen3.5-2B 33.7%
  77. 77 Llama 3.1-8B 23.7%
  78. 78 Gemma 3 4B 22.8%
  79. 79 Qwen3.5-0.8B 18.9%
  80. 80 GPT-4.1 nano 15%

Sources: Official MultiChallenge leaderboard and labs’ model cards

#EntryScore ± 95% CISettingDateSource
1 GPT-6.1 Sol 82% ±4.2 Sep 2026 Official leaderboard
2 Gemini 4 Hydrogen 80.6% ±5 Nov 2026 Official leaderboard
3 Gemini 4 Argon 78.7% ±5.3 Sep 2026 Official leaderboard
4 Muse Spark 75.5% ±4 Apr 2026 Official leaderboard
5 Muse Spark 1.1 75.3% ±0.6 Jul 2026 Official leaderboard
6 DeepSeek V4.1 Flash 1017 74.9% ±5.3 Oct 2026 Official leaderboard
7 Muse Spark Open 74.2% ±5.6 Oct 2026 Official leaderboard
8 GPT-6.1 Terra 74.2% ±5.4 Oct 2026 Official leaderboard
9 GPT-6.2 Astra 73.2% ±5.8 Nov 2026 Official leaderboard
10 GPT-6.2 Sol 72% ±5.8 Nov 2026 Official leaderboard
11 Gemini 3.1 Pro 71.4% ±1.7 Feb 2026 Official leaderboard
12 GPT-5.4 Pro 69.2% ±3.1 Mar 2026 Official leaderboard
13 Claude Haiku 5.5 67.7% ±5.3 Oct 2026 Official leaderboard
14 Gemini 3 Pro 65.7% ±2.2 Nov 2025 Official leaderboard
15 GPT-5.1 63.4% ±2.1 thinking Nov 2025 Official leaderboard
16 GPT-5 63.2% ±1.6 thinking Aug 2025 Official leaderboard
17 o3-pro 62.4% ±2.5 high Jun 2025 Official leaderboard
18 Kimi K2.5 61.4% ±0.5 Jan 2026 Official leaderboard
19 Gemini 3.1 Flash-Lite 60.6% ±0.4 Mar 2026 Official leaderboard
20 GPT-5 mini 59% ±4.1 thinking Aug 2025 Official leaderboard
21 Claude Opus 4.5 59% ±0.4 thinking Nov 2025 Official leaderboard
22 Claude Opus 4 58.6% ±0.8 thinking May 2025 Official leaderboard
23 Claude Opus 4.1 57.2% ±1 thinking Aug 2025 Official leaderboard
24 Claude Sonnet 4 57.1% ±0.8 thinking May 2025 Official leaderboard
25 o3 56.6% ±3.3 high Apr 2025 Official leaderboard
26 Nemotron 3.5 Bolt E4B 56% ±6.2 thinking Oct 2026 Official leaderboard
27 Claude Opus 4.6 56% ±1.1 non-thinking Feb 2026 Official leaderboard
28 Kimi K2 Thinking 55.4% ±1.9 Nov 2025 Official leaderboard
29 Claude Sonnet 4.5 55.3% ±1.5 thinking Sep 2025 Official leaderboard
30 Gemini 2.5 Pro Jun 2025 53.6% ±1.3 Jun 2025 Official leaderboard
31 Claude 3.7 Sonnet 51.6% ±3.1 thinking Feb 2025 Official leaderboard
32 GPT-5.1 Instant 51.2% ±1.8 Nov 2025 Official leaderboard
33 Claude Haiku 4.5 50.5% ±2 thinking Oct 2025 Official leaderboard
34 DeepSeek-V3.1 46.1% ±1.1 Aug 2025 Official leaderboard
35 gpt-oss-120b 45.3% ±2.2 Aug 2025 Official leaderboard
36 o4-mini 44.9% ±0.9 high Apr 2025 Official leaderboard
37 Gemma 4 E4B 43.2% ±6.1 thinking Mar 2026 Official leaderboard
38 Qwen3-235B-A22B 41.2% ±1.8 Apr 2025 Official leaderboard
39 GPT-4.1 39.4% ±0.9 Apr 2025 Official leaderboard
40 Nemotron 3 Nano 4B 38.7% ±6.3 thinking Mar 2026 Official leaderboard
41 Gemini 2.0 Flash Feb 2025 36.4% ±0.8 Feb 2025 Official leaderboard
42 Gemma 4 E2B 36% ±5.9 thinking Mar 2026 Official leaderboard
43 Ministral 3 8B 27.8% ±5.7 thinking Official leaderboard
44 Ministral 3 3B 25.7% ±5.4 thinking Official leaderboard
45 Qwen3.5-4B 25.1% ±5.3 thinking Feb 2026 Official leaderboard
46 Llama 3.1-8B 23.7% ±5.6 Jul 2024 Official leaderboard
47 Gemma 3 4B 22.8% ±5.5 Mar 2025 Official leaderboard
#ModelScore ± 95% CISettingDateSource
1 GPT-6.1 Sol 82% ±4.2 Sep 2026 Official leaderboard
2 Gemini 4 Hydrogen 80.6% ±5 Nov 2026 Official leaderboard
3 Gemini 4 Argon 78.7% ±5.3 Sep 2026 Official leaderboard
4 Muse Spark 75.5% ±4 Apr 2026 Official leaderboard
5 Muse Spark 1.1 75.3% ±0.6 Jul 2026 Official leaderboard
6 DeepSeek V4.1 Flash 1017 74.9% ±5.3 Oct 2026 Official leaderboard
7 Muse Spark Open 74.2% ±5.6 Oct 2026 Official leaderboard
8 GPT-6.1 Terra 74.2% ±5.4 Oct 2026 Official leaderboard
9 GPT-6.2 Astra 73.2% ±5.8 Nov 2026 Official leaderboard
10 GPT-6.2 Sol 72% ±5.8 Nov 2026 Official leaderboard
11 Gemini 3.1 Pro 71.4% ±1.7 Feb 2026 Official leaderboard
12 GPT-5.4 Pro 69.2% ±3.1 Mar 2026 Official leaderboard
13 Claude Haiku 5.5 67.7% ±5.3 Oct 2026 Official leaderboard
14 Gemini 3 Pro 65.7% ±2.2 Nov 2025 Official leaderboard
15 GPT-5.1 63.4% ±2.1 thinking Nov 2025 Official leaderboard
16 GPT-5 63.2% ±1.6 thinking Aug 2025 Official leaderboard
17 o3-pro 62.4% ±2.5 high Jun 2025 Official leaderboard
18 Kimi K2.5 61.4% ±0.5 Jan 2026 Official leaderboard
19 Gemini 3.1 Flash-Lite 60.6% ±0.4 Mar 2026 Official leaderboard
20 GPT-5 mini 59% ±4.1 thinking Aug 2025 Official leaderboard
21 Claude Opus 4.5 59% ±0.4 thinking Nov 2025 Official leaderboard
22 Claude Opus 4 58.6% ±0.8 thinking May 2025 Official leaderboard
23 Claude Opus 4.1 57.2% ±1 thinking Aug 2025 Official leaderboard
24 Claude Sonnet 4 57.1% ±0.8 thinking May 2025 Official leaderboard
25 o3 56.6% ±3.3 high Apr 2025 Official leaderboard
26 Nemotron 3.5 Bolt E4B 56% ±6.2 thinking Oct 2026 Official leaderboard
27 Claude Opus 4.6 56% ±1.1 non-thinking Feb 2026 Official leaderboard
28 Kimi K2 Thinking 55.4% ±1.9 Nov 2025 Official leaderboard
29 Claude Sonnet 4.5 55.3% ±1.5 thinking Sep 2025 Official leaderboard
30 Gemini 2.5 Pro Jun 2025 53.6% ±1.3 Jun 2025 Official leaderboard
31 Claude 3.7 Sonnet 51.6% ±3.1 thinking Feb 2025 Official leaderboard
32 GPT-5.1 Instant 51.2% ±1.8 Nov 2025 Official leaderboard
33 Claude Haiku 4.5 50.5% ±2 thinking Oct 2025 Official leaderboard
34 DeepSeek-V3.1 46.1% ±1.1 Aug 2025 Official leaderboard
35 gpt-oss-120b 45.3% ±2.2 Aug 2025 Official leaderboard
36 o4-mini 44.9% ±0.9 high Apr 2025 Official leaderboard
37 Gemma 4 E4B 43.2% ±6.1 thinking Mar 2026 Official leaderboard
38 Qwen3-235B-A22B 41.2% ±1.8 Apr 2025 Official leaderboard
39 GPT-4.1 39.4% ±0.9 Apr 2025 Official leaderboard
40 Nemotron 3 Nano 4B 38.7% ±6.3 thinking Mar 2026 Official leaderboard
41 Gemini 2.0 Flash Feb 2025 36.4% ±0.8 Feb 2025 Official leaderboard
42 Gemma 4 E2B 36% ±5.9 thinking Mar 2026 Official leaderboard
43 Qwen3.5-4B 25.1% ±5.3 thinking Feb 2026 Official leaderboard
44 Llama 3.1-8B 23.7% ±5.6 Jul 2024 Official leaderboard
45 Gemma 3 4B 22.8% ±5.5 Mar 2025 Official leaderboard
#ModelScoreSettingDateSource
1 GPT-6.1 Sol 82% Sep 2026 Official leaderboard
2 Gemini 4 Hydrogen 80.6% Nov 2026 Official leaderboard
3 Gemini 4 Argon 78.7% Sep 2026 Official leaderboard
4 Nova 2.0 Pro Preview 77.7% Dec 2025 Lab model card
5 Nova 2 Lite 76.6% Dec 2025 Lab model card
6 Muse Spark 75.5% Apr 2026 Official leaderboard
7 Nova 2 Omni Preview 75.5% Dec 2025 Lab model card
8 Muse Spark 1.1 75.3% Jul 2026 Official leaderboard
9 DeepSeek V4.1 Flash 1017 74.9% Oct 2026 Official leaderboard
10 Muse Spark Open 74.2% Oct 2026 Official leaderboard
11 GPT-6.1 Terra 74.2% Oct 2026 Official leaderboard
12 GPT-6.2 Astra 73.2% Nov 2026 Official leaderboard
13 GPT-6.2 Sol 72% Nov 2026 Official leaderboard
14 Gemini 3.1 Pro 71.4% Feb 2026 Official leaderboard
15 Seed2.0 Lite (0428) 69.9% Apr 2026 Lab model card
16 Ling-3.1-flash 69.8% Oct 2026 Lab model card
17 GPT-5 69.6% Aug 2025 Lab model card
18 GPT-5.4 Pro 69.2% Mar 2026 Official leaderboard
19 Seed2.0 Pro 68.3% Feb 2026 Lab model card
20 Claude Haiku 5.5 67.7% Oct 2026 Official leaderboard
21 Qwen3.5 397B-A17B 67.6% Feb 2026 Lab model card
22 Seed1.8 66.7% Dec 2025 Lab model card
23 ERNIE 5.0 66% Feb 2026 Lab model card
24 Gemini 3 Pro 65.7% Nov 2025 Official leaderboard
25 GLM-5 64.8% Feb 2026 Lab model card
26 Nemotron 3 Ultra 63.8% Jun 2026 Lab model card
27 GPT-5.1 63.4% thinking Nov 2025 Official leaderboard
28 Qwen3-Max-2026-01-23 63.3% Jan 2026 Lab model card
29 STEP3-VL-10B 62.6% Jan 2026 Lab model card
30 Tencent HY 2.0 Think 62.6% Dec 2025 Lab model card
31 o3-pro 62.4% high Jun 2025 Official leaderboard
32 GPT-5 mini 62.3% Aug 2025 Lab model card
33 Qwen3.5-122B-A10B 61.5% Feb 2026 Lab model card
34 Kimi K2.5 61.4% Jan 2026 Official leaderboard
35 Solar Open 2 61% Jul 2026 Lab model card
36 Qwen3.5-27B 60.8% Feb 2026 Lab model card
37 Gemini 3.1 Flash-Lite 60.6% Mar 2026 Official leaderboard
38 Mercury 3 60.4% Oct 2026 Lab model card
39 Qwen3.5-35B-A3B 60% Feb 2026 Lab model card
40 Claude Opus 4.5 59% thinking Nov 2025 Official leaderboard
41 Claude Opus 4 58.6% thinking May 2025 Official leaderboard
42 Claude Opus 4.1 57.2% thinking Aug 2025 Official leaderboard
43 Claude Sonnet 4 57.1% thinking May 2025 Official leaderboard
44 o3 56.6% high Apr 2025 Official leaderboard
45 Nemotron 3.5 Bolt E4B 56% thinking Oct 2026 Official leaderboard
46 Claude Opus 4.6 56% non-thinking Feb 2026 Official leaderboard
47 Step 3.5 Flash 55.7% Feb 2026 Lab model card
48 Kimi K2 Thinking 55.4% Nov 2025 Official leaderboard
49 Claude Sonnet 4.5 55.3% thinking Sep 2025 Official leaderboard
50 Nemotron 3 Super 55.2% Mar 2026 Lab model card
51 GPT-5 nano 54.9% Aug 2025 Lab model card
52 Qwen3.5-9B 54.5% Mar 2026 Lab model card
53 Kimi K2 Jul 2025 54.1% Jul 2025 Lab model card
54 Gemini 2.5 Pro Jun 2025 53.6% Jun 2025 Official leaderboard
55 MAI-Thinking-1 53% Jun 2026 Lab model card
56 GLM-4.5 52.8% Aug 2025 Lab model card
57 Claude 3.7 Sonnet 51.6% thinking Feb 2025 Official leaderboard
58 GPT-5.1 Instant 51.2% Nov 2025 Official leaderboard
59 Ring-1T 50.9% Oct 2025 Lab model card
60 Claude Haiku 4.5 50.5% thinking Oct 2025 Official leaderboard
61 Qwen3.5-4B 49% Mar 2026 Lab model card
62 DeepSeek-V3.1 46.1% Aug 2025 Official leaderboard
63 gpt-oss-120b 45.3% Aug 2025 Official leaderboard
64 Nemotron-Cascade 2 30B-A3B 45.3% Mar 2026 Lab model card
65 o4-mini 44.9% high Apr 2025 Official leaderboard
66 MiniMax-M1-80k 44.7% Jun 2025 Lab model card
67 Gemma 4 E4B 43.2% thinking Mar 2026 Official leaderboard
68 GLM-4.5-Air 42.5% Aug 2025 Lab model card
69 Qwen3-235B-A22B 41.2% Apr 2025 Official leaderboard
70 GPT-4.1 39.4% Apr 2025 Official leaderboard
71 Nemotron 3 Nano 4B 38.7% thinking Mar 2026 Official leaderboard
72 Nemotron 3 Nano 30B-A3B 38.5% Dec 2025 Lab model card
73 Gemini 2.0 Flash Feb 2025 36.4% Feb 2025 Official leaderboard
74 Gemma 4 E2B 36% thinking Mar 2026 Official leaderboard
75 GPT-4.1 mini 35.8% Apr 2025 Lab model card
76 Qwen3.5-2B 33.7% Mar 2026 Lab model card
77 Llama 3.1-8B 23.7% Jul 2024 Official leaderboard
78 Gemma 3 4B 22.8% Mar 2025 Official leaderboard
79 Qwen3.5-0.8B 18.9% Mar 2026 Lab model card
80 GPT-4.1 nano 15% Apr 2025 Lab model card

Labs report scores for 42 models in their cards. Switch the scores above to rank them together.

12 models have held number one on the official leaderboard since its release in January 2025 9 models have held number one with claims included since its release in January 2025

Lab claims count from the day the card came out, independent runs from the model’s release. We draw one row each for the benchmark’s own leaderboard and what labs claim in their cards.

Meta Google DeepMind Anthropic OpenAI Amazon
Own leaderboard 12 leaders
Lab claims 4 leaders

Sources: Official MultiChallenge leaderboard and labs’ model cards

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

Plot with other benchmarks
On the benchmark’s own leaderboard Reported by the lab in its model card Best so far
0% 20% 40% 60% 80% 100% 2025 2026 2027 Near the ceiling (85%) GPT-6.1 Sol Nova 2.0 Pro Preview GPT-5 o3-pro o3 Claude 3.7 Sonnet Gemini 2.0 Flash 0% 20% 40% 60% 80% 100% 2025 2026 2027 Near the ceiling (85%) GPT-6.1 Sol GPT-5

Sources: Official MultiChallenge leaderboard and labs’ model cards

5 labs put it in a launch post in the last six months

Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.

OpenAI
MiniMax
Moonshot AI
Amazon
Tencent
NVIDIA
ByteDance Seed
Qwen
Baidu
Microsoft
Upstage
Ant Group
Inception Labs
DateLaunch postScoreAs the post writes it
Oct 23, 2026 Nemotron 3.5 Bolt E4B 55.6 MultiChallenge
Oct 19, 2026 Mercury 3 60.4 MultiChallenge
Oct 13, 2026 Ling-3.1-flash 69.8 MultiChallenge
Jul 22, 2026 Solar Open 2 61 Multi-Challenge
Jun 4, 2026 Nemotron 3 Ultra 63.8 Multi-Challenge
Jun 2, 2026 MAI-Thinking-1 53 Multi-Challenge
Apr 28, 2026 Seed2.0 Lite (0428) 69.9 MultiChallenge
Mar 19, 2026 Nemotron-Cascade 2 30B-A3B 45.3 Scale AI Multi-Challenge
Mar 11, 2026 Nemotron 3 Super 55.2 Scale AI Multi-Challenge
Mar 2, 2026 Qwen3.5-0.8B 18.9 MultiChallenge
Mar 2, 2026 Qwen3.5-2B 33.7 MultiChallenge
Mar 2, 2026 Qwen3.5-4B 49 MultiChallenge
Mar 2, 2026 Qwen3.5-9B 54.5 MultiChallenge
Feb 24, 2026 Qwen3.5-35B-A3B 60 MultiChallenge
Feb 24, 2026 Qwen3.5-27B 60.8 MultiChallenge
Feb 24, 2026 Qwen3.5-122B-A10B 61.5 MultiChallenge
Feb 16, 2026 Qwen3.5 397B-A17B 67.6 MultiChallenge
Feb 14, 2026 Seed2.0 Pro 68.3 MultiChallenge
Feb 6, 2026 ERNIE 5.0 66 MultiChallenge
Jan 23, 2026 Qwen3-Max-2026-01-23 63.3 MultiChallenge
Dec 18, 2025 Seed1.8 66.7 MultiChallenge
Dec 15, 2025 Nemotron 3 Nano 30B-A3B 38.5 Scale AI Multi Challenge
Dec 5, 2025 Tencent HY 2.0 Think 62.6 Multi Challenge
Dec 2, 2025 Nova 2 Omni Preview 75.5 MultiChallenge acc
Dec 2, 2025 Nova 2 Lite 76.6 MultiChallenge acc
Dec 2, 2025 Nova 2.0 Pro Preview 77.7 MultiChallenge acc
Aug 7, 2025 GPT-5 nano 54.9 Scale MultiChallenge (o3-mini grader)
Aug 7, 2025 GPT-5 mini 62.3 Scale MultiChallenge (o3-mini grader)
Aug 7, 2025 GPT-5 69.6 Scale MultiChallenge** Multi-turn instruction following
Jul 11, 2025 Kimi K2 54.1 Multi-Challenge, Acc
Jun 16, 2025 MiniMax-M1-80k 44.7 MultiChallenge
Apr 16, 2025 o4-mini 43 Scale MultiChallenge
Apr 16, 2025 o3 56.5 Scale MultiChallenge Multi-turn instruction following
Apr 14, 2025 GPT-4.1 nano 15 MultiChallenge (GPT-4o grader) 15.0 (o3-mini grader 31.1)
Apr 14, 2025 GPT-4.1 mini 35.8 MultiChallenge (GPT-4o grader) 35.8 (o3-mini grader 42.2)
Apr 14, 2025 GPT-4.1 38.3 MultiChallenge

21 system cards report it, from 11 labs

Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.

DateDocumentScoreAs the document writes itSection
Oct 23, 2026 Nemotron 3.5 Bolt E4B technical report 55.6 MultiChallenge, released Q4_K_M imatrix checkpoint (Table 5). BF16 53.1 (Table 3) Capabilities
Jul 22, 2026 Solar Open 2 technical report 61 Multi-Challenge (Table 3) Capabilities
Jun 9, 2026 Nemotron 3 Ultra technical report 63.8 Multi-Challenge (Table 10) Capabilities
Jun 2, 2026 MAI-Thinking-1 technical report 53 Multi-Challenge, Gemini 2.5 Pro judge (Table 12, rounded) Capabilities
Mar 21, 2026 Seed1.8 model card 66.7 MultiChallenge (Table 1) Capabilities
Mar 19, 2026 Nemotron-Cascade 2 30B-A3B technical report 45.3 Scale AI Multi-Challenge (Table 1) Capabilities
Mar 11, 2026 Nemotron 3 Super technical report 55.2 Scale AI Multi-Challenge (Table 5) Capabilities
Feb 17, 2026 GLM-5 technical report 64.8 Multi Challenge, Instruction Following panel (Figure 11) Capabilities
Feb 14, 2026 Seed2.0 Pro model card 68.3 MultiChallenge (Table 3) Capabilities
Feb 11, 2026 Step 3.5 Flash technical report 55.7 MultiChallenge, o3-mini judge (Table 5). PaCoRe 60.8 Capabilities
Feb 4, 2026 ERNIE 5.0 technical report 66 MultiChallenge (Table 2) Capabilities
Jan 14, 2026 STEP3-VL-10B technical report 62.6 MultiChallenge, SeRe (Table 2) Capabilities
Dec 23, 2025 Nemotron 3 Nano 30B-A3B technical report 38.5 Scale AI Multi Challenge (Table 3) Capabilities
Dec 2, 2025 Nova 2 Omni Preview technical report 75.5 MultiChallenge acc (Table 5) Capabilities
Dec 2, 2025 Nova 2 Lite technical report 76.6 MultiChallenge acc (Table 1) Capabilities
Dec 2, 2025 Nova 2.0 Pro Preview technical report 77.7 MultiChallenge acc (Table 1) Capabilities
Oct 21, 2025 Ring-1T technical report 50.9 MultiChallenge (Table 1) Capabilities
Aug 8, 2025 GLM-4.5 technical report 52.8 MultiChallenge (Table 6) Capabilities
Aug 8, 2025 GLM-4.5-Air technical report 42.5 MultiChallenge (Table 6) Capabilities
Jul 21, 2025 Kimi K2 technical report 54.1 Multi-Challenge (Acc.) (Table 3) Capabilities
Jun 16, 2025 MiniMax-M1-80k technical report 44.7 MultiChallenge, GPT-4o judge (Table 2) Capabilities

The lab's number comes out ahead in 3 of 6 pairs that differ

Each score a lab reported for its own model, beside the same model's best score on the leaderboard. Gaps mostly come from different harnesses, effort levels and task subsets. 1 of them match, which usually means one side copied the other. Every benchmark.

Model Lab says Board says Gap
Qwen3.5-4B 49% MultiChallenge 25.1% Leaderboard, thinking +24 pts
GPT-5 69.6% Scale MultiChallenge** Multi-turn instruction following 63.2% Leaderboard, thinking +6.4 pts
GPT-5 mini 62.3% Scale MultiChallenge (o3-mini grader) 59% Leaderboard, thinking +3.3 pts
o4-mini 43% Scale MultiChallenge 44.9% Leaderboard, high −1.9 pts
GPT-4.1 38.3% MultiChallenge 39.4% Leaderboard −1.1 pts
Nemotron 3.5 Bolt E4B 55.6% MultiChallenge 56% Leaderboard, thinking −0.4 pts
o3 56.5% Scale MultiChallenge Multi-turn instruction following 56.6% Leaderboard, high Same

More in Instructions and chat

All 144 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%
  • 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%
  • ComplexConstraints In use Models answer 75 expert-written professional prompts whose constraints depend on each other, fire conditionally or must be inferred. An LLM judge grades 1,559 rubric criteria one at a time, and a task passes only if all criteria pass. GPT-6 Astra 70.8%

Data audit

11 of 12 checks pass, 1 warning. Look first at the Lab claims check. 1 of 7 pairs differs by more than 10 points, the largest for Qwen3.5-4B (+24 pts). Every benchmark

Catalog record 2 pass
  • Pass Task access Public tasks, plus a private test set.
  • Pass Confidence Checked against a primary source.
Scores 5 pass
  • Pass Independent runs Scores from the official leaderboard (47 entries), launch posts (36 scores) and system cards (21 scores).
  • Pass Models scored 80 models we track, plus 2 other entries on its leaderboard.
  • Pass Newest score The newest is from November 2026, for GPT-6.2 Astra.
  • Pass Frontier models 4 of the top 10 have a score, led by GPT-6.2 Astra (#1).
  • Pass Headroom The best is 82%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on labs.scale.com: 46 rows for models we track and 2 other entries at the last read.
  • Pass Last read November 19th, 2026.
  • Pass Read worked The last read worked.
Consistency 1 warning 1 pass
  • Warning Lab claims 1 of 7 pairs differs by more than 10 points, the largest for Qwen3.5-4B (+24 pts).
  • Pass Known issues No published audit found problems.

How we check