Benchmark Almanac

AIME 2026

GPT-5.5 100% leads since April 2026
GPT-5.5 100% leads since April 2026
In use

Models solve the 30 integer-answer problems from the 2026 AIME I and II math competitions. Scored by exact answer match.

An exam, not a dedicated benchmark. Top models score above 90%.

Released
February 2026
Kind
Competition problems with answers
Level
Competition
Scored
Checked automatically
Answers
Short answer
Tasks
All public
Strongest model tested
GPT-5.5 #20 on the capability index
Last in a launch post
October 2026
Last model added
October 2026
Scores refreshed
November 12th, 2026
31
Models scored on MathArena’s leaderboard
34
Launch posts list it, from 17 labs
23
System cards report it, from 14 labs

2 models share the top score of 100% among models released by July 2026

Best score per model from MathArena’s leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart.

GPT-5.5 leads with 100%, 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.

The newest model with a score here came out in July 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

31 models. Scroll sideways for the rest.

  1. 100.0 GPT-5.5
  2. 100.0 Claude Opus 4.8
  3. 99.2 GPT-5.4
  4. 98.3 GPT-5.2
  5. 98.3 Gemini 3.1 Pro
  6. 96.7 Gemini 3 Flash
  7. 96.7 Step 3.5 Flash
  8. 96.7 Claude Opus 4.6
  9. 96.7 GLM-5
  10. 96.7 DeepSeek-V4-Pro
  11. 96.7 Kimi K3
  12. 96.7 Gemini 3.6 Flash
  13. 95.8 Kimi K2.5
  14. 95.8 GLM-5.1
  15. 95.8 Claude Opus 4.7
  16. 95.8 Kimi K2.6
  17. 95.8 DeepSeek-V4-Flash
  18. 95.0 Gemini 3.5 Flash
  19. 95.0 Step 3.7 Flash
  20. 94.2 Grok 4.1 Fast
  21. 94.2 DeepSeek-V3.2
  22. 94.2 Qwen3.5 397B-A17B
  23. 93.3 Qwen3.5-35B-A3B
  24. 92.5 Qwen3.5-9B
  25. 91.7 Gemini 3 Pro
  26. 91.7 Qwen3.5-27B
  27. 91.7 Nemotron 3 Super
  28. 90.0 GLM-5.2
  29. 88.3 Qwen3-30B-A3B-Thinking
  30. 82.5 Qwen3-4B-Thinking-2507
  31. 82.5 QED-Nano
  1. 1 GPT-5.5 xhigh 100%
  2. 2 Claude Opus 4.8 max 100%
  3. 3 GPT-5.4 xhigh 99.2%
  4. 4 GPT-5.2 high 98.3%
  5. 5 Gemini 3.1 Pro high 98.3%
  6. 6 Gemini 3 Flash high 96.7%
  7. 7 Step 3.5 Flash 96.7%
  8. 8 Claude Opus 4.6 high 96.7%
  9. 9 GLM-5 96.7%
  10. 10 DeepSeek-V4-Pro max 96.7%
  11. 11 Kimi K3 max 96.7%
  12. 12 Gemini 3.6 Flash 96.7%
  13. 13 Kimi K2.5 high 95.8%
  14. 14 GLM-5.1 95.8%
  15. 15 Claude Opus 4.7 xhigh 95.8%
  16. 16 Kimi K2.6 high 95.8%
  17. 17 DeepSeek-V4-Flash max 95.8%
  18. 18 Gemini 3.5 Flash high 95%
  19. 19 Step 3.7 Flash high 95%
  20. 20 Grok 4.1 Fast reasoning 94.2%
  21. 21 DeepSeek-V3.2 thinking 94.2%
  22. 22 Qwen3.5 397B-A17B 94.2%
  23. 23 Qwen3.5-35B-A3B 93.3%
  24. 24 Qwen3.5-9B 92.5%
  25. 25 Gemini 3 Pro 91.7%
  26. 26 Qwen3.5-27B 91.7%
  27. 27 Nemotron 3 Super 91.7%
  28. 28 GLM-5.2 max 90%
  29. 29 Qwen3-30B-A3B-Thinking Jul 2025 88.3%
  30. 30 Qwen3-4B-Thinking-2507 82.5%
  31. 31 QED-Nano 82.5%

Source: MathArena’s leaderboard

64 models. Scroll sideways for the rest.

  1. 100.0 GPT-5.5
  2. 100.0 Claude Opus 4.8
  3. 100.0 DeepSeek V4.1 Flash
  4. 99.6 ERNIE 5.1
  5. 99.2 GLM-5.2
  6. 99.2 GPT-5.4
  7. 98.3 GPT-5.2
  8. 98.3 Gemini 3.1 Pro
  9. 97.8 Beam 1.0
  10. 97.1 Inkling
  11. 96.7 Gemini 3 Flash
  12. 96.7 Step 3.5 Flash
  13. 96.7 Claude Opus 4.6
  14. 96.7 GLM-5
  15. 96.7 DeepSeek-V4-Pro
  16. 96.7 Kimi K3
  17. 96.7 Gemini 3.6 Flash
  18. 96.4 Kimi K2.6
  19. 95.8 Kimi K2.5
  20. 95.8 GLM-5.1
  21. 95.8 Claude Opus 4.7
  22. 95.8 DeepSeek-V4-Flash
  23. 95.8 Ring-2.6-1T
  24. 95.7 Solar Open 2
  25. 95.5 Inkling-Small
  26. 95.3 Qwen 3.6 Plus
  27. 95.3 Solar Pro 4
  28. 95.0 Ring-2.5-1T
  29. 95.0 Gemini 3.5 Flash
  30. 95.0 Step 3.7 Flash
  31. 94.7 Muse Glimmer
  32. 94.5 MAI-Thinking-1
  33. 94.2 Seed2.0 Pro
  34. 94.2 MiniMax-M2.7
  35. 94.2 Grok 4.1 Fast
  36. 94.2 DeepSeek-V3.2
  37. 94.2 Qwen3.5 397B-A17B
  38. 94.1 Qwen3.6 27B
  39. 93.3 Qwen3.5-35B-A3B
  40. 93.3 Keye-VL-2.0-30B-A3B
  41. 92.7 Qwen 3.6 35B-A3B
  42. 92.6 EXAONE 4.5 33B
  43. 92.5 Qwen3.5-9B
  44. 92.5 MAI-Code-1-Flash
  45. 92.3 K-EXAONE 2.0
  46. 91.7 Gemini 3 Pro
  47. 91.7 Qwen3.5-27B
  48. 91.7 Nemotron 3 Super
  49. 91.2 Spark-X2.6-4B
  50. 90.9 Nemotron-Cascade 2 30B-A3B
  51. 89.2 Gemma 4 31B IT
  52. 88.3 Qwen3-30B-A3B-Thinking
  53. 88.3 Gemma 4 26B A4B
  54. 87.4 Ling-2.6-1T
  55. 87.2 MiniMax-M2.5
  56. 83.0 Solar Pro 3 (March 2026 update)
  57. 82.5 Qwen3-4B-Thinking-2507
  58. 82.5 QED-Nano
  59. 77.5 Gemma 4 12B
  60. 69.1 DiffusionGemma 26B A4B
  61. 63.7 Beam 1.0 Base
  62. 50.0 LFM2.5-8B-A1B
  63. 42.5 Gemma 4 E4B
  64. 37.5 Gemma 4 E2B
  1. 1 GPT-5.5 xhigh 100%
  2. 2 Claude Opus 4.8 max 100%
  3. 3 DeepSeek V4.1 Flash 100%
  4. 4 ERNIE 5.1 99.6%
  5. 5 GLM-5.2 99.2%
  6. 6 GPT-5.4 xhigh 99.2%
  7. 7 GPT-5.2 high 98.3%
  8. 8 Gemini 3.1 Pro high 98.3%
  9. 9 Beam 1.0 97.8%
  10. 10 Inkling 97.1%
  11. 11 Gemini 3 Flash high 96.7%
  12. 12 Step 3.5 Flash 96.7%
  13. 13 Claude Opus 4.6 high 96.7%
  14. 14 GLM-5 96.7%
  15. 15 DeepSeek-V4-Pro max 96.7%
  16. 16 Kimi K3 max 96.7%
  17. 17 Gemini 3.6 Flash 96.7%
  18. 18 Kimi K2.6 96.4%
  19. 19 Kimi K2.5 high 95.8%
  20. 20 GLM-5.1 95.8%
  21. 21 Claude Opus 4.7 xhigh 95.8%
  22. 22 DeepSeek-V4-Flash max 95.8%
  23. 23 Ring-2.6-1T 95.8%
  24. 24 Solar Open 2 95.7%
  25. 25 Inkling-Small 95.5%
  26. 26 Qwen 3.6 Plus 95.3%
  27. 27 Solar Pro 4 95.3%
  28. 28 Ring-2.5-1T 95%
  29. 29 Gemini 3.5 Flash high 95%
  30. 30 Step 3.7 Flash high 95%
  31. 31 Muse Glimmer 94.7%
  32. 32 MAI-Thinking-1 94.5%
  33. 33 Seed2.0 Pro 94.2%
  34. 34 MiniMax-M2.7 94.2%
  35. 35 Grok 4.1 Fast reasoning 94.2%
  36. 36 DeepSeek-V3.2 thinking 94.2%
  37. 37 Qwen3.5 397B-A17B 94.2%
  38. 38 Qwen3.6 27B 94.1%
  39. 39 Qwen3.5-35B-A3B 93.3%
  40. 40 Keye-VL-2.0-30B-A3B 93.3%
  41. 41 Qwen 3.6 35B-A3B 92.7%
  42. 42 EXAONE 4.5 33B 92.6%
  43. 43 Qwen3.5-9B 92.5%
  44. 44 MAI-Code-1-Flash 92.5%
  45. 45 K-EXAONE 2.0 92.3%
  46. 46 Gemini 3 Pro 91.7%
  47. 47 Qwen3.5-27B 91.7%
  48. 48 Nemotron 3 Super 91.7%
  49. 49 Spark-X2.6-4B 91.2%
  50. 50 Nemotron-Cascade 2 30B-A3B 90.9%
  51. 51 Gemma 4 31B IT 89.2%
  52. 52 Qwen3-30B-A3B-Thinking Jul 2025 88.3%
  53. 53 Gemma 4 26B A4B 88.3%
  54. 54 Ling-2.6-1T 87.4%
  55. 55 MiniMax-M2.5 87.2%
  56. 56 Solar Pro 3 (March 2026 update) 83%
  57. 57 Qwen3-4B-Thinking-2507 82.5%
  58. 58 QED-Nano 82.5%
  59. 59 Gemma 4 12B 77.5%
  60. 60 DiffusionGemma 26B A4B 69.1%
  61. 61 Beam 1.0 Base 63.7%
  62. 62 LFM2.5-8B-A1B 50%
  63. 63 Gemma 4 E4B 42.5%
  64. 64 Gemma 4 E2B 37.5%

Sources: MathArena’s leaderboard and labs’ model cards

#ModelScoreSettingDateSource
1 GPT-5.5 100% xhigh Apr 2026 Official leaderboard
2 Claude Opus 4.8 100% max May 2026 Official leaderboard
3 GPT-5.4 99.2% xhigh Mar 2026 Official leaderboard
4 GPT-5.2 98.3% high Dec 2025 Official leaderboard
5 Gemini 3.1 Pro 98.3% high Feb 2026 Official leaderboard
6 Gemini 3 Flash 96.7% high Dec 2025 Official leaderboard
7 Step 3.5 Flash 96.7% Feb 2026 Official leaderboard
8 Claude Opus 4.6 96.7% high Feb 2026 Official leaderboard
9 GLM-5 96.7% Feb 2026 Official leaderboard
10 DeepSeek-V4-Pro 96.7% max Apr 2026 Official leaderboard
11 Kimi K3 96.7% max Jul 2026 Official leaderboard
12 Gemini 3.6 Flash 96.7% Jul 2026 Official leaderboard
13 Kimi K2.5 95.8% high Jan 2026 Official leaderboard
14 GLM-5.1 95.8% Apr 2026 Official leaderboard
15 Claude Opus 4.7 95.8% xhigh Apr 2026 Official leaderboard
16 Kimi K2.6 95.8% high Apr 2026 Official leaderboard
17 DeepSeek-V4-Flash 95.8% max Apr 2026 Official leaderboard
18 Gemini 3.5 Flash 95% high May 2026 Official leaderboard
19 Step 3.7 Flash 95% high May 2026 Official leaderboard
20 Grok 4.1 Fast 94.2% reasoning Nov 2025 Official leaderboard
21 DeepSeek-V3.2 94.2% thinking Dec 2025 Official leaderboard
22 Qwen3.5 397B-A17B 94.2% Feb 2026 Official leaderboard
23 Qwen3.5-35B-A3B 93.3% Feb 2026 Official leaderboard
24 Qwen3.5-9B 92.5% Feb 2026 Official leaderboard
25 Gemini 3 Pro 91.7% Nov 2025 Official leaderboard
26 Qwen3.5-27B 91.7% Feb 2026 Official leaderboard
27 Nemotron 3 Super 91.7% Mar 2026 Official leaderboard
28 GLM-5.2 90% max Jun 2026 Official leaderboard
29 Qwen3-30B-A3B-Thinking Jul 2025 88.3% Jul 2025 Official leaderboard
30 Qwen3-4B-Thinking-2507 82.5% Aug 2025 Official leaderboard
31 QED-Nano 82.5% Feb 2026 Official leaderboard
#ModelScoreSettingDateSource
1 GPT-5.5 100% xhigh Apr 2026 Official leaderboard
2 Claude Opus 4.8 100% max May 2026 Official leaderboard
3 DeepSeek V4.1 Flash 100% Sep 2026 Lab model card
4 ERNIE 5.1 99.6% May 2026 Lab model card
5 GLM-5.2 99.2% Jun 2026 Lab model card
6 GPT-5.4 99.2% xhigh Mar 2026 Official leaderboard
7 GPT-5.2 98.3% high Dec 2025 Official leaderboard
8 Gemini 3.1 Pro 98.3% high Feb 2026 Official leaderboard
9 Beam 1.0 97.8% Oct 2026 Lab model card
10 Inkling 97.1% Jul 2026 Lab model card
11 Gemini 3 Flash 96.7% high Dec 2025 Official leaderboard
12 Step 3.5 Flash 96.7% Feb 2026 Official leaderboard
13 Claude Opus 4.6 96.7% high Feb 2026 Official leaderboard
14 GLM-5 96.7% Feb 2026 Official leaderboard
15 DeepSeek-V4-Pro 96.7% max Apr 2026 Official leaderboard
16 Kimi K3 96.7% max Jul 2026 Official leaderboard
17 Gemini 3.6 Flash 96.7% Jul 2026 Official leaderboard
18 Kimi K2.6 96.4% Apr 2026 Lab model card
19 Kimi K2.5 95.8% high Jan 2026 Official leaderboard
20 GLM-5.1 95.8% Apr 2026 Official leaderboard
21 Claude Opus 4.7 95.8% xhigh Apr 2026 Official leaderboard
22 DeepSeek-V4-Flash 95.8% max Apr 2026 Official leaderboard
23 Ring-2.6-1T 95.8% May 2026 Lab model card
24 Solar Open 2 95.7% Jul 2026 Lab model card
25 Inkling-Small 95.5% Jul 2026 Lab model card
26 Qwen 3.6 Plus 95.3% Apr 2026 Lab model card
27 Solar Pro 4 95.3% Aug 2026 Lab model card
28 Ring-2.5-1T 95% Feb 2026 Lab model card
29 Gemini 3.5 Flash 95% high May 2026 Official leaderboard
30 Step 3.7 Flash 95% high May 2026 Official leaderboard
31 Muse Glimmer 94.7% Aug 2026 Lab model card
32 MAI-Thinking-1 94.5% Jun 2026 Lab model card
33 Seed2.0 Pro 94.2% Feb 2026 Lab model card
34 MiniMax-M2.7 94.2% May 2026 Lab model card
35 Grok 4.1 Fast 94.2% reasoning Nov 2025 Official leaderboard
36 DeepSeek-V3.2 94.2% thinking Dec 2025 Official leaderboard
37 Qwen3.5 397B-A17B 94.2% Feb 2026 Official leaderboard
38 Qwen3.6 27B 94.1% Apr 2026 Lab model card
39 Qwen3.5-35B-A3B 93.3% Feb 2026 Official leaderboard
40 Keye-VL-2.0-30B-A3B 93.3% May 2026 Lab model card
41 Qwen 3.6 35B-A3B 92.7% Apr 2026 Lab model card
42 EXAONE 4.5 33B 92.6% Apr 2026 Lab model card
43 Qwen3.5-9B 92.5% Feb 2026 Official leaderboard
44 MAI-Code-1-Flash 92.5% Jun 2026 Lab model card
45 K-EXAONE 2.0 92.3% Jul 2026 Lab model card
46 Gemini 3 Pro 91.7% Nov 2025 Official leaderboard
47 Qwen3.5-27B 91.7% Feb 2026 Official leaderboard
48 Nemotron 3 Super 91.7% Mar 2026 Official leaderboard
49 Spark-X2.6-4B 91.2% Oct 2026 Lab model card
50 Nemotron-Cascade 2 30B-A3B 90.9% Mar 2026 Lab model card
51 Gemma 4 31B IT 89.2% Apr 2026 Lab model card
52 Qwen3-30B-A3B-Thinking Jul 2025 88.3% Jul 2025 Official leaderboard
53 Gemma 4 26B A4B 88.3% Apr 2026 Lab model card
54 Ling-2.6-1T 87.4% Apr 2026 Lab model card
55 MiniMax-M2.5 87.2% May 2026 Lab model card
56 Solar Pro 3 (March 2026 update) 83% Mar 2026 Lab model card
57 Qwen3-4B-Thinking-2507 82.5% Aug 2025 Official leaderboard
58 QED-Nano 82.5% Feb 2026 Official leaderboard
59 Gemma 4 12B 77.5% Jun 2026 Lab model card
60 DiffusionGemma 26B A4B 69.1% Jun 2026 Lab model card
61 Beam 1.0 Base 63.7% Oct 2026 Lab model card
62 LFM2.5-8B-A1B 50% May 2026 Lab model card
63 Gemma 4 E4B 42.5% Apr 2026 Lab model card
64 Gemma 4 E2B 37.5% Apr 2026 Lab model card

Labs report scores for 38 models in their cards. Switch the scores above to rank them together.

The best score, 100%, costs $0.16 a task

Runs on the cost frontier score higher than any run that's cheaper. DeepSeek-V4-Flash at max comes within 4.2 points of the best for $0.009 a task, 18 times cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.

xAI OpenAI StepFun DeepSeek Cost frontier One model at each effort level
0% 20% 40% 60% 80% 100% $0.001 $0.01 $0.10 $1.00 Cost per task, log scale GPT-5.5, xhigh Step 3.5 Flash Grok 4.1 Fast, reasoning 0% 20% 40% 60% 80% 100% $0.001 $0.01 $0.10 $1.00 Cost per task, log scale GPT-5.5, xhigh Step 3.5 Flash

Source: MathArena’s leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
GPT-5.5 xhigh 100% $0.16 5,219
Claude Opus 4.8 max 100% $0.54 21,360
GPT-5.4 xhigh 99.2% $0.16 10,743
GPT-5.2 high 98.3% $0.12 8,403
Gemini 3.1 Pro high 98.3% $0.17 14,364
Gemini 3 Flash high 96.7% $0.06 20,504
Step 3.5 Flash 96.7% $0.01 42,072
Claude Opus 4.6 high 96.7% $0.33 13,329
GLM-5 96.7% $0.07 23,541
DeepSeek-V4-Pro max 96.7% $0.08 23,567
Kimi K3 max 96.7% $0.11 7,031
Gemini 3.6 Flash 96.7% $0.08 10,951
Kimi K2.5 high 95.8% $0.07 22,849
GLM-5.1 95.8% $0.09 26,546
Claude Opus 4.7 xhigh 95.8% $0.27 10,728
Kimi K2.6 high 95.8% $0.09 22,722
DeepSeek-V4-Flash max 95.8% $0.009 33,588
Gemini 3.5 Flash high 95% $0.13 13,992
Step 3.7 Flash high 95% $0.01 50,365
Grok 4.1 Fast reasoning 94.2% $0.005 9,618
DeepSeek-V3.2 thinking 94.2% $0.006 14,854
Qwen3.5 397B-A17B 94.2% $0.07 20,797
Qwen3.5-35B-A3B 93.3% $0.04 31,856
Qwen3.5-9B 92.5% $0.005 31,184
Gemini 3 Pro 91.7% $0.18 14,712
Qwen3.5-27B 91.7% $0.05 29,363
Nemotron 3 Super 91.7% 24,114
GLM-5.2 max 90% $0.08 18,424
Qwen3-30B-A3B-Thinking Jul 2025 88.3% $0.005 17,939
Qwen3-4B-Thinking-2507 82.5% $0.006 21,206
QED-Nano 82.5% 35,191

3 models have held number one on MathArena’s leaderboard since its release in February 2026 3 models have held number one with claims included since its release in February 2026

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

OpenAI Ant Group Qwen Moonshot AI Baidu DeepSeek
MathArena leaderboard Passed 85% in Feb 2026, by GPT-5.2
GPT-5.5 100%
Lab claims Passed 85% in Feb 2026, by Ring-2.5-1T
DeepSeek V4.1 Flash 100%

Sources: MathArena’s leaderboard and labs’ model cards

Qwen3-30B-A3B-Thinking on the official leaderboard passed 85% in July 2025

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

On MathArena’s leaderboard Reported by the lab in its model card Best so far
0% 20% 40% 60% 80% 100% 2026 2027 Near the ceiling (85%) GPT-5.5 GPT-5.2 Grok 4.1 Fast Qwen3-30B-A3B-Thinking 0% 20% 40% 60% 80% 100% 2026 2027 Near the ceiling (85%) GPT-5.5 Gemini 3 Pro

Sources: MathArena’s leaderboard and labs’ model cards

9 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.

Ant Group
Z.ai
ByteDance Seed
Qwen
NVIDIA
Upstage
Google DeepMind
LG AI Research
Moonshot AI
Baidu
Kuaishou
Liquid AI
Microsoft
Thinking Machines
Meta
Reflection
iFlytek
DateLaunch postScoreAs the post writes it
Oct 22, 2026 Spark-X2.6-4B 91.2 AIME 2026
Oct 5, 2026 Beam 1.0 97.8 AIME 2026
Aug 11, 2026 Solar Pro 4 95.3 AIME 2026* (in-house)
Aug 10, 2026 Muse Glimmer 94.7 AIME 2026
Jul 31, 2026 K-EXAONE 2.0 92.3 AIME 2026
Jul 30, 2026 Inkling-Small 95.5 AIME 2026
Jul 22, 2026 Solar Open 2 95.7 AIME2026 (MathArena)
Jul 15, 2026 Inkling 97.1 AIME 2026
Jun 16, 2026 GLM-5.2 99.2 AIME 2026
Jun 10, 2026 DiffusionGemma 26B A4B 69.1 AIME 2026 no tools
Jun 3, 2026 Gemma 4 12B 77.5 AIME 2026 no tools
Jun 2, 2026 MAI-Code-1-Flash 92.5 AIME 2026
Jun 2, 2026 MAI-Thinking-1 94.5 AIME 2026
May 28, 2026 LFM2.5-8B-A1B 50 AIME26
May 25, 2026 Keye-VL-2.0-30B-A3B 93.3 AIME2026
May 14, 2026 Ring-2.6-1T 95.8 AIME 26 (xhigh)
May 9, 2026 ERNIE 5.1 99.6 Best of AIME26 w/tools (avg@32) 99.6 / AIME26 (avg@32) 95.0
Apr 29, 2026 Ling-2.6-1T 87.4 AIME26 (non-reasoning, chart)
Apr 22, 2026 Qwen3.6 27B 94.1 AIME26 (full AIME 2026 I and II)
Apr 20, 2026 Kimi K2.6 96.4 AIME 2026
Apr 15, 2026 Qwen 3.6 35B-A3B 92.7 AIME26 (full AIME 2026 I and II)
Apr 9, 2026 EXAONE 4.5 33B 92.6 AIME 2026
Apr 7, 2026 GLM-5.1 95.3 AIME 2026
Apr 2, 2026 Gemma 4 E4B 42.5 AIME 2026 no tools
Apr 2, 2026 Gemma 4 E2B 37.5 AIME 2026 no tools
Apr 2, 2026 Gemma 4 26B A4B 88.3 AIME 2026 no tools
Apr 2, 2026 Qwen 3.6 Plus 95.3 AIME26
Apr 2, 2026 Gemma 4 31B IT 89.2 AIME 2026 no tools
Mar 24, 2026 Solar Pro 3 (March 2026 update) 83 AIME'26
Mar 19, 2026 Nemotron-Cascade 2 30B-A3B 90.9 AIME 2026 (TIR 95.0)
Feb 16, 2026 Qwen3.5 397B-A17B 91.3 AIME26
Feb 14, 2026 Seed2.0 Pro 94.2 AIME 2026
Feb 11, 2026 GLM-5 92.7 AIME 2026 I
Feb 10, 2026 Ring-2.5-1T 95 Best of AIME 26 (Mean@64) 93.75 / Heavy Thinking 95.00

23 system cards report it, from 14 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 19, 2026 Muse Glimmer technical report 94.7 AIME 2026, high, 93.0 xhigh (Tables 2 and 5) Capabilities
Oct 5, 2026 Beam 1.0 Base technical report 63.7 AIME 2026, 0-shot temperature 0.7, Table 23 Capabilities
Sep 10, 2026 DeepSeek V4.1 Flash technical report 100 AIME 2026 (Pass@1) at reasoning effort 100 (Appendix B.3 text, Figure 12) Capabilities
Aug 5, 2026 K-EXAONE 2.0 technical report 92.3 AIME 2026 (Table 6, Reasoning mode) Capabilities
Jul 31, 2026 DiffusionGemma 26B A4B technical report 69.1 AIME 2026, TD with thinking, 50.8 no-think, 84.2 AR mode (Table 3) Capabilities
Jul 30, 2026 Inkling-Small model card 95.5 AIME 2026 (5. Evaluations table) Capabilities
Jul 24, 2026 Gemma 4 E4B technical report 42.5 AIME 2026 no tools, thinking (Table 5) Capabilities
Jul 24, 2026 Gemma 4 E2B technical report 37.5 AIME 2026 no tools, thinking (Table 5) Capabilities
Jul 24, 2026 Gemma 4 31B IT technical report 89.2 AIME 2026 no tools, thinking (Table 5) Capabilities
Jul 24, 2026 Gemma 4 26B A4B technical report 88.3 AIME 2026 no tools, thinking (Table 5) Capabilities
Jul 22, 2026 Solar Open 2 technical report 95.7 AIME2026, MathArena set (Table 3) Capabilities
Jul 15, 2026 Inkling model card 97.1 AIME 2026 (5. Evaluations table, effort 0.99) Capabilities
Jun 13, 2026 Ring-2.6-1T technical report 95.8 AIME 2026 (Avg@64), xhigh (Table 6). High gives 87.86 Capabilities
Jun 13, 2026 Ling-2.6-1T technical report 87.4 AIME26, Mean@64, CoT (Table 5) Capabilities
Jun 9, 2026 Keye-VL-2.0-30B-A3B technical report 93.3 AIME2026 (Figure 5, printed value) Capabilities
Jun 2, 2026 MAI-Thinking-1 technical report 94.5 AIME 2026, pass@1, 256k max output tokens (Table 11) Capabilities
Jun 2, 2026 MAI-Code-1-Flash model card 92.5 AIME 2026, Acc (Core reasoning table) Capabilities
May 26, 2026 MiniMax-M2.7 technical report 94.2 AIME 2026, AA Index v4.0 protocol, pass@1 (Table 4) Capabilities
May 26, 2026 MiniMax-M2.5 technical report 87.2 AIME 2026, AA Index v4.0 protocol, pass@1 (Table 4) Capabilities
Apr 9, 2026 EXAONE 4.5 33B technical report 92.6 AIME 2026 (Table 3, Reasoning mode) Capabilities
Mar 19, 2026 Nemotron-Cascade 2 30B-A3B technical report 90.9 AIME 2026, no tools (Table 1). TIR 95.0 Capabilities
Feb 17, 2026 GLM-5 technical report 92.7 AIME 2026 I (Table 7) Capabilities
Feb 14, 2026 Seed2.0 Pro model card 94.2 AIME 2026 (Table 3) Capabilities

The lab's number comes out ahead in 2 of 5 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. Every benchmark.

Model Lab says Board says Gap
GLM-5.2 99.2% AIME 2026 90% Leaderboard, max +9.2 pts
GLM-5 92.7% AIME 2026 I 96.7% Leaderboard −4.0 pts
Qwen3.5 397B-A17B 91.3% AIME26 94.2% Leaderboard −2.9 pts
Kimi K2.6 96.4% AIME 2026 95.8% Leaderboard, high +0.6 pts
GLM-5.1 95.3% AIME 2026 95.8% Leaderboard −0.5 pts

Versions

We switch the top row to each new version when it comes out. Open it for one row per version. An old version stays in play while people still test it, so whoever first takes it past 85% gets credit for closing it out.

OpenAI DeepSeek xAI
AIME 2026 Passed 85% in Feb 2026, by GPT-5.2
GPT-5.5 100%
AIME 2024 Passed 85% in Jan 2025, by o3-mini
gpt-oss-120b 96.6%
AIME 2025 Passed 85% in Feb 2025, by Grok 3
Grok 4 Heavy 100%
AIME 2026 Passed 85% in Feb 2026, by GPT-5.2
GPT-5.5 100%

Sources: Official leaderboards and labs’ model cards

Replaces AIME 2025

More in Math

All 113 in Math
  • IMO-AnswerBench In use 400 olympiad problems with short verifiable answers, robustified against memorization. Scored by answer match. GLM-5.2 91%
  • HMMT 2026 In use Models solve the 33 final-answer problems from the February 2026 Harvard-MIT Mathematics Tournament. Scored by exact answer match. GPT-5.5 98.5%
  • PolyMath In use Models solve 9,000 parallel math problems in 18 languages across four difficulty levels, from school math to frontier math. Final answers are checked. Qwen3.7-Max 86.5%
  • MathArena (Apex, IMO, USAMO) In use Evaluates models on newly released math competitions after release. Final-answer or human/LLM-judged proof scores. Claude Opus 4.8 81.3%
  • MathArena Apex Shortlist In use Models solve 47 hard final-answer problems that MathArena selected from 2025 competitions, where the best models scored about 50% at selection. Scored by accuracy over repeated attempts. GPT-5.5 98.4%
  • FrontierMath Tiers 1-3 v2 In use Models solve 295 unpublished math problems from undergraduate to advanced graduate level and submit answers that code checks automatically. Scored by share correct. Claude Fable 5.2 95.1%