Skip to main content

Vals AIME (AIME)

Challenging national math exam given to top high-school students

Data verified

How BenchLM shows AIME

BenchLM mirrors the public Vals AI AIME leaderboard captured from https://www.vals.ai/benchmarks/aime and updated by Vals on April 16, 2026. The snapshot preserves overall scores, uncertainty, latency, cost-per-test metadata, and task-level scores where Vals publishes them.

AIME is display only on BenchLM. Vals proprietary or Vals-hosted aggregate views are useful context, but BenchLM does not use them as weighted ranking inputs or as a replacement for benchmark-native source records.

96 Vals rows3 task viewspublic datasetTasks: Overall, AIME 2024, AIME 2025Display only

AIME score on AIME — April 16, 2026

BenchLM mirrors the published aime score view for AIME. Gemini 3.1 Pro Preview leads the public snapshot at 98.13% , followed by GPT-5.2 (96.88%) and Muse Spark (96.88%). BenchLM does not use these results to rank models overall.

96 modelsExternal benchmark mirrorsCurrentDisplay onlyUpdated April 16, 2026

The published AIME snapshot is tightly clustered at the top: Gemini 3.1 Pro Preview sits at 98.13%, while the third row is only 1.25 points behind. The broader top-10 spread is 2.50 points, so many of the published scores sit in a relatively narrow band.

96 models have been evaluated on AIME. The benchmark falls in the External benchmark mirrors category. BenchLM tracks this category separately from its weighted global scoring system, so these results are best compared on the dedicated Korean benchmark views. AIME is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About AIME

Year

2026

Tasks

AIME math problems

Format

Accuracy score

Difficulty

Competition math

BenchLM mirrors the public Vals AI AIME leaderboard as display-only external evidence. The captured snapshot preserves overall scores, task-level scores where Vals publishes them, uncertainty, latency, and cost-per-test metadata. It is excluded from BenchLM weighted rankings.

BenchLM freshness & provenance

Version

AIME 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

AIME score table (96 models)

1
Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview
98.13%
2
GPT-5.2openai/gpt-5.2-2025-12-11
96.88%
3
Muse Sparkmeta/muse_spark
96.88%
4
Gemini 3 Pro Previewgoogle/gemini-3-pro-preview
96.68%
5
GPT-5.4openai/gpt-5.4-2026-03-05
96.67%
6
Grok 4.20 0309 Reasoninggrok/grok-4.20-0309-reasoning
96.46%
7
Claude Opus 4.7anthropic/claude-opus-4-7
96.25%
8
Gemini 3 Flash Previewgoogle/gemini-3-flash-preview
95.63%
9
Claude Opus 4.6 Thinkinganthropic/claude-opus-4-6-thinking
95.63%
10
GPT-5.4 Miniopenai/gpt-5.4-mini-2026-03-17
95.63%
11
Kimi K2.5 Thinkingkimi/kimi-k2.5-thinking
95.63%
12
Claude Opus 4.5 20251101 Thinkinganthropic/claude-opus-4-5-20251101-thinking
95.42%
13
Qwen3.6 Plusalibaba/qwen3.6-plus
94.58%
14
GPT-5openai/gpt-5-2025-08-07
93.37%
15
GPT-5.1openai/gpt-5.1-2025-11-13
93.33%
16
GLM 4.7zai/glm-4.7
93.33%
17
GLM 4.6zai/glm-4.6
92.71%
18
GPT Oss 120bfireworks/gpt-oss-120b
92.60%
19
Qwen3.5 Flashalibaba/qwen3.5-flash
92.50%
20
Claude Sonnet 4.6anthropic/claude-sonnet-4-6
92.29%
21
Grok 4.1 Fast Reasoninggrok/grok-4-1-fast-reasoning
91.88%
22
GLM 5.1zai/glm-5.1
91.88%
23
GLM 5 Thinkingzai/glm-5-thinking
91.67%
24
GPT-5 Miniopenai/gpt-5-mini-2025-08-07
91.46%
25
Grok 4 Fast Reasoninggrok/grok-4-fast-reasoning
91.25%
26
MiniMax M2.7minimax/MiniMax-M2.7
91.04%
27
Grok 4 0709grok/grok-4-0709
90.56%
28
GPT-5.4 Nanoopenai/gpt-5.4-nano-2026-03-17
88.75%
29
MiniMax M2.5minimax/MiniMax-M2.5
88.75%
30
Claude Sonnet 4.5 20250929 Thinkinganthropic/claude-sonnet-4-5-20250929-thinking
88.19%
31
GLM 4.5zai/glm-4.5
86.67%
32
O3 Miniopenai/o3-mini-2025-01-31
86.46%
33
Qwen3.5 Plus Thinkingalibaba/qwen3.5-plus-thinking
86.04%
34
GPT Oss 20bfireworks/gpt-oss-20b
86.04%
35
Gemini 2.5 Pro Exp 03 25google/gemini-2.5-pro-exp-03-25
85.83%
36
Kimi K2 Thinkingkimi/kimi-k2-thinking
85.42%
37
O3openai/o3-2025-04-16
85.28%
38
Grok 3 Mini Fast High Reasoninggrok/grok-3-mini-fast-high-reasoning
85.00%
39
DeepSeek V3p2 Thinkingfireworks/deepseek-v3p2-thinking
84.58%
40
Qwen3 235b A22bfireworks/qwen3-235b-a22b
83.96%
41
O4 Miniopenai/o4-mini-2025-04-16
83.67%
42
Magistral Medium 2509mistralai/magistral-medium-2509
83.54%
43
Gemini 3.1 Flash Lite Previewgoogle/gemini-3.1-flash-lite-preview
83.33%
44
Claude Haiku 4.5 20251001 Thinkinganthropic/claude-haiku-4-5-20251001-thinking
82.71%
45
GPT-5 Nanoopenai/gpt-5-nano-2025-08-07
81.18%
46
Qwen3 Maxalibaba/qwen3-max
81.04%
47
Magistral Small 2509mistralai/magistral-small-2509
80.68%
48
Claude Opus 4.1 20250805 Thinkinganthropic/claude-opus-4-1-20250805-thinking
78.18%
49
MiniMax M2.1minimax/MiniMax-M2.1
77.92%
50
Claude Opus 4.5anthropic/claude-opus-4-5-20251101
76.88%
51
Claude Sonnet 4 20250514 Thinkinganthropic/claude-sonnet-4-20250514-thinking
76.25%
52
DeepSeek R1fireworks/deepseek-r1
73.96%
53
O1openai/o1-2024-12-17
71.46%
54
Grok 3 Mini Fast Low Reasoninggrok/grok-3-mini-fast-low-reasoning
70.63%
55
DeepSeek V3p2fireworks/deepseek-v3p2
64.79%
56
Moonshotai Kimi K2 Instructtogether/moonshotai/Kimi-K2-Instruct
62.71%
57
Qwen3 Max Previewalibaba/qwen3-max-preview
60.69%
58
Grok 3grok/grok-3
58.75%
59
Langston Nim Nvidia Llama 3.3 Nemotron Super 49b V1 42e84561 Thinkingtogether/langston/nim/nvidia/llama-3.3-nemotron-super-49b-v1-42e84561-thinking
53.54%
60
DeepSeek V3 0324fireworks/deepseek-v3-0324
52.20%
61
Gemini 2.5 Flash Preview 09 2025 Thinkinggoogle/gemini-2.5-flash-preview-09-2025-thinking
51.46%
62
Gemini 2.5 Flash Preview 09 2025google/gemini-2.5-flash-preview-09-2025
49.79%
63
GPT-4.1 Miniopenai/gpt-4.1-mini-2025-04-14
49.38%
64
Claude 3.7 Sonnet 20250219 Thinkinganthropic/claude-3-7-sonnet-20250219-thinking
44.58%
65
Claude Opus 4.1anthropic/claude-opus-4-1-20250805
44.24%
66
Mistral Large 2512mistralai/mistral-large-2512
42.92%
67
Mistral Medium 2505mistralai/mistral-medium-2505
42.29%
68
Gemini 2.5 Flash Lite Preview 09 2025 Thinkinggoogle/gemini-2.5-flash-lite-preview-09-2025-thinking
42.08%
69
Claude Opus 4anthropic/claude-opus-4-20250514
41.25%
70
GPT-4.1openai/gpt-4.1-2025-04-14
39.58%
71
Claude Sonnet 4anthropic/claude-sonnet-4-20250514
38.54%
72
Grok 4 Fast Non Reasoninggrok/grok-4-fast-non-reasoning
33.33%
73
Gemini 2.0 Flash 001google/gemini-2.0-flash-001
29.79%
74
DeepSeek V3fireworks/deepseek-v3
27.50%
75
Grok 4.1 Fast Non Reasoninggrok/grok-4-1-fast-non-reasoning
27.08%
76
GPT-4.1 Nanoopenai/gpt-4.1-nano-2025-04-14
26.46%
77
Gemini 2.5 Flash Lite Preview 09 2025google/gemini-2.5-flash-lite-preview-09-2025
26.25%
78
Llama4 Maverick Instruct Basicfireworks/llama4-maverick-instruct-basic
25.21%
79
Claude 3.7 Sonnetanthropic/claude-3-7-sonnet-20250219
22.29%
80
Meta Llama Llama 4 Scout 17B 16E Instructtogether/meta-llama/Llama-4-Scout-17B-16E-Instruct
18.96%
81
Gemini 1.5 Pro 002google/gemini-1.5-pro-002
18.75%
82
Gemini 1.5 Flash 002google/gemini-1.5-flash-002
17.29%
83
Meta Llama Llama 3.3 70B Instruct Turbotogether/meta-llama/Llama-3.3-70B-Instruct-Turbo
16.04%
84
Grok 2 1212grok/grok-2-1212
15.21%
85
GPT-4oopenai/gpt-4o-2024-08-06
13.96%
86
Command A 03 2025cohere/command-a-03-2025
13.33%
87
GPT-4oopenai/gpt-4o-2024-11-20
11.88%
88
GPT-4o Miniopenai/gpt-4o-mini-2024-07-18
11.46%
89
Claude 3.5 Sonnetanthropic/claude-3-5-sonnet-20241022
10.00%
90
Langston Nim Nvidia Llama 3.3 Nemotron Super 49b V1 42e84561together/langston/nim/nvidia/llama-3.3-nemotron-super-49b-v1-42e84561
9.38%
91
Mistral Large 2411mistralai/mistral-large-2411
9.17%
92
Mistral Small 2402mistralai/mistral-small-2402
5.63%
93
Mistral Small 2503mistralai/mistral-small-2503
3.54%
94
Claude 3.5 Haikuanthropic/claude-3-5-haiku-20241022
3.33%
95
Jamba Mini 1.6ai21labs/jamba-mini-1.6
0.42%
96
Jamba Large 1.6ai21labs/jamba-large-1.6
0.42%

FAQ

What does AIME measure?

Challenging national math exam given to top high-school students

Which model leads the published AIME snapshot?

Gemini 3.1 Pro Preview currently leads the published AIME snapshot with 98.13% aime score. BenchLM shows this benchmark for display only and does not use it in overall rankings.

How many models are evaluated on AIME?

96 AI models are included in BenchLM's mirrored AIME snapshot, based on the public leaderboard captured on April 16, 2026.

Last updated: April 16, 2026 · mirrored from the public benchmark leaderboard

The AI models change fast. We track them for you.

For engineers, researchers, and the plain curious — a weekly brief on new models, ranking shifts, and pricing changes.

Free. No spam. Unsubscribe anytime.