FrontierMath Tiers 1–3 (v2)Benchmark scores and sources

Published result; benchmark version and evaluation conditions remain in the id and result note.

pillar: math · weight 1 within pillar · unit: % (higher is better) · official board ↗
Row Model Result Normalized (0–100) Confidence
1 GPT-6 Astra OpenAI 93.7%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Aug 30, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		93.7 100% confidence 100 percent, Full
2 GPT-6.1 Sol OpenAI 93.7%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Sep 29, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		93.7 100% confidence 100 percent, Full
3 Claude Opus 5.5 Anthropic 91.2%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Sep 22, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		91.2 100% confidence 100 percent, Full
4 Claude Fable 5.1 Anthropic 90.2%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Sep 1, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		90.2 100% confidence 100 percent, Full
5 GPT-6 Sol OpenAI 89.8%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Sep 22, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		89.8 100% confidence 100 percent, Full
6 GPT-5.6 Sol OpenAI 89.1%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Jul 9, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		89.1 100% confidence 100 percent, Full
7 Claude Sonnet 5.5 Anthropic 88.8%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Sep 29, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		88.8 93% confidence 93 percent, High
8 GPT-5.5 Pro OpenAI 87.7%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] xhighPublished Jun 12, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		87.7 53% confidence 53 percent, Medium
9 Claude Fable 5 Anthropic 87.0%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Jun 9, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		87.0 100% confidence 100 percent, Full
10 GPT-5.6 Terra OpenAI 86.0%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Jul 9, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		86.0 100% confidence 100 percent, Full
11 Claude Opus 5 Anthropic 85.6%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Jul 24, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		85.6 100% confidence 100 percent, Full
12 GPT-5.5 OpenAI 85.3%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] xhighPublished Jun 11, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		85.3 100% confidence 100 percent, Full
13 GPT-5.4 Pro OpenAI 82.5%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] xhighPublished Jun 13, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		82.5 69% confidence 69 percent, Medium
14 GPT-5.6 Luna OpenAI 82.1%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Jul 9, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		82.1 100% confidence 100 percent, Full
15 Claude Opus 4.8 Anthropic 80%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Jun 10, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		80.0 100% confidence 100 percent, Full
16 GPT-6 Luna OpenAI 78.9%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Sep 22, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		78.9 100% confidence 100 percent, Full
17 GPT-5.4 OpenAI 78.6%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] xhighPublished Jun 11, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		78.6 100% confidence 100 percent, Full
18 Qwen3.8 Max Alibaba / Qwen 74.7%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] xhighPublished Aug 4, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		74.7 85% confidence 85 percent, High
19 Muse Spark 1.3 Meta 74.4%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] xhighPublished Sep 16, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		74.4 93% confidence 93 percent, High
20 Muse Spark 1.3 Meta 74.0%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Sep 18, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		74.0 93% confidence 93 percent, High
21 GPT-5.2 Pro OpenAI 74%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] xhighPublished Jun 13, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		74.0 48% confidence 48 percent, Low
22 Kimi K3 Moonshot AI 72.2%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Jul 17, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		72.2 100% confidence 100 percent, Full
23 Gemini 3.7 Flash Google 71.6%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] highPublished Aug 14, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		71.6 100% confidence 100 percent, Full
24 Claude Opus 4.7 Anthropic 70.2%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Jun 10, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		70.2 100% confidence 100 percent, Full
25 GLM-5.3 Z.ai 68.8%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Aug 25, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		68.8 93% confidence 93 percent, High
26 Gemini 3.8 Flash Google 68.4%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] highPublished Sep 2, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		68.4 100% confidence 100 percent, Full
27 GPT-5.2 OpenAI 67.4%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] xhighPublished Jun 11, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		67.4 100% confidence 100 percent, Full
28 Claude Opus 4.6 Anthropic 66.0%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Jun 11, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		66.0 100% confidence 100 percent, Full
29 Grok 4.6 xAI 66.0%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] xhighPublished Aug 14, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		66.0 100% confidence 100 percent, Full
30 Qwen3.8 Max 0902 Alibaba / Qwen 65.6%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] xhighPublished Sep 2, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		65.6 40% confidence 40 percent, Low
31 Claude Sonnet 5 Anthropic 65.6%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Jun 30, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		65.6 93% confidence 93 percent, High
32 Qwen3.7 Max Alibaba / Qwen 64.6%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Jun 13, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		64.6 53% confidence 53 percent, Medium
33 DeepSeek V4 Pro 0813 DeepSeek 64.6%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Aug 19, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		64.6 69% confidence 69 percent, Medium
34 Gemini 3.5 Flash Google 62.8%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] highPublished Jun 10, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		62.8 100% confidence 100 percent, Full
35 Gemini 3.1 Pro Preview Google 59.6%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Jun 11, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		59.6 100% confidence 100 percent, Full
36 GLM-5.2 Z.ai 59.2%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Jun 19, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		59.2 100% confidence 100 percent, Full
37 Gemini 3.6 Flash Google 58.9%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] highPublished Aug 2, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		58.9 100% confidence 100 percent, Full
38 DeepSeek V4 Flash 0731 DeepSeek 57.5%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Aug 2, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		57.5 69% confidence 69 percent, Medium
39 Kimi K2.6 Moonshot AI 57.2%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Jun 10, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		57.2 85% confidence 85 percent, High
40 Grok 4.5 xAI 57.2%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] highPublished Jul 9, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		57.2 100% confidence 100 percent, Full
41 GPT-5 Pro OpenAI 55.8%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] highPublished Jun 12, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		55.8 64% confidence 64 percent, Medium
42 GLM-5.3-Flash Z.ai 55.8%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Aug 27, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		55.8 100% confidence 100 percent, Full
43 GPT-5 OpenAI 55.4%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] highPublished Jun 10, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		55.4 100% confidence 100 percent, Full
44 GLM-5.2 Z.ai 54.7%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] lowPublished Aug 29, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		54.7 100% confidence 100 percent, Full
45 Kimi K2.7 Code Moonshot AI 54.0%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Jun 13, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		54.0 48% confidence 48 percent, Low
46 Grok 4.7 xAI 53.0%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] xhighPublished Sep 22, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		53.0 100% confidence 100 percent, Full
47 Gemini 3 Flash Preview Google 51.2%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Jun 11, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		51.2 93% confidence 93 percent, High
48 GPT-5.4 mini OpenAI 51.2%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] xhighPublished Jun 12, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		51.2 100% confidence 100 percent, Full
49 GPT-5 Mini OpenAI 46.7%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] highPublished Jun 12, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		46.7 99% confidence 99 percent, High
50 Inkling Small Thinking Machines Lab 46.3%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] xhighPublished Aug 14, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		46.3 100% confidence 100 percent, Full
51 DeepSeek V4 Pro DeepSeek 45.3%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Jun 17, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		45.3 85% confidence 85 percent, High
52 GPT-5.4 nano OpenAI 44.9%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] highPublished Jun 12, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		44.9 100% confidence 100 percent, Full
53 Grok 4.20 (Reasoning) xAI 44.9%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Jul 13, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		44.9 80% confidence 80 percent, High
54 Grok 4.3 xAI 42.8%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] highPublished Jun 17, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		42.8 80% confidence 80 percent, High
55 GLM-5.2 Z.ai 42.5%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] nonePublished Aug 29, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		42.5 100% confidence 100 percent, Full
56 GPT-5.6 Luna OpenAI 41.4%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] lowPublished Aug 29, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		41.4 100% confidence 100 percent, Full
57 GPT-5.6 Luna OpenAI 39.6%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] nonePublished Aug 29, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		39.6 100% confidence 100 percent, Full
58 Qwen3.6 Plus Alibaba / Qwen 38.2%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Aug 30, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		38.2 80% confidence 80 percent, High
59 GPT-5 OpenAI 37.2%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] lowPublished Aug 27, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		37.2 100% confidence 100 percent, Full
60 GLM-5.1 Z.ai 36.8%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Aug 29, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		36.8 64% confidence 64 percent, Medium
61 o4-mini OpenAI 36.1%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] highPublished Jun 11, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		36.1 87% confidence 87 percent, High
62 Qwen3.6 27B Alibaba / Qwen 35.1%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Aug 28, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		35.1 53% confidence 53 percent, Medium
63 Qwen3.7 Plus Alibaba / Qwen 34.4%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] nonePublished Aug 29, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		34.4 64% confidence 64 percent, Medium
64 Claude Opus 4.5 Anthropic 34.4%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] 32KPublished Jun 11, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		34.4 100% confidence 100 percent, Full
65 Qwen3.6 27B Alibaba / Qwen 34.0%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] nonePublished Aug 29, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		34.0 53% confidence 53 percent, Medium
66 o3 OpenAI 33.3%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] highPublished Aug 27, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		33.3 87% confidence 87 percent, High
67 Inkling Thinking Machines Lab 33.3%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] xhighPublished Aug 6, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		33.3 100% confidence 100 percent, Full
68 Qwen3.6 Plus Alibaba / Qwen 32.3%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] nonePublished Aug 29, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		32.3 80% confidence 80 percent, High
69 Qwen3.5 397B-A17B Alibaba / Qwen 31.2%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] nonePublished Aug 30, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		31.2 69% confidence 69 percent, Medium
70 o3 OpenAI 29.8%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] mediumPublished Aug 27, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		29.8 87% confidence 87 percent, High
71 Qwen3.5 397B-A17B Alibaba / Qwen 29.5%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Aug 28, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		29.5 69% confidence 69 percent, Medium
72 o4-mini OpenAI 28.8%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] mediumPublished Aug 27, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		28.8 87% confidence 87 percent, High
73 Gemini 3.1 Flash Lite Google 27.7%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] highPublished Aug 30, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		27.7 32% confidence 32 percent, Low
74 GPT-5.5 Instant OpenAI 26.3%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Aug 2, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		26.3 64% confidence 64 percent, Medium
75 Gemini 3.5 Flash Lite Google 26.0%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] highPublished Aug 2, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		26.0 100% confidence 100 percent, Full
76 GLM-5.1 Z.ai 24.9%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] nonePublished Aug 28, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		24.9 64% confidence 64 percent, Medium
77 Gemini 2.5 Pro Google 24.6%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Jun 11, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		24.6 90% confidence 90 percent, High
78 GPT-5.4 mini OpenAI 24.6%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] lowPublished Aug 28, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		24.6 100% confidence 100 percent, Full
79 Claude Sonnet 4.5 Anthropic 23.9%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] 32KPublished Jun 11, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		23.9 80% confidence 80 percent, High
80 Qwen3.6 Flash Alibaba / Qwen 22.5%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] nonePublished Aug 29, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		22.5 32% confidence 32 percent, Low
81 Gemini 3.1 Flash Lite Google 22.5%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] lowPublished Aug 30, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		22.5 32% confidence 32 percent, Low
82 Gemini 3.1 Flash Lite Google 21.4%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] minimalPublished Aug 28, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		21.4 32% confidence 32 percent, Low
83 Qwen3.6 35B-A3B Alibaba / Qwen 20.4%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] nonePublished Aug 30, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		20.4 37% confidence 37 percent, Low
84 GPT-5.4 nano OpenAI 20.4%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] lowPublished Aug 28, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		20.4 100% confidence 100 percent, Full
85 GPT-5 Nano OpenAI 20%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] highPublished Jun 12, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		20.0 85% confidence 85 percent, High
86 Qwen3.7 Flash Alibaba / Qwen 19.3%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Aug 29, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		19.3 32% confidence 32 percent, Low
87 Qwen3.7 Flash Alibaba / Qwen 19.3%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] nonePublished Aug 29, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		19.3 32% confidence 32 percent, Low
88 o3 OpenAI 19.3%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] lowPublished Aug 27, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		19.3 87% confidence 87 percent, High
89 Qwen3 Max Alibaba / Qwen 18.9%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Aug 30, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		18.9 69% confidence 69 percent, Medium
90 Qwen3.5 Flash Alibaba / Qwen 18.2%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] nonePublished Aug 28, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		18.2 64% confidence 64 percent, Medium
91 GPT-5 OpenAI 18.2%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] minimalPublished Aug 27, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		18.2 100% confidence 100 percent, Full
92 GPT-5 Mini OpenAI 18.2%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] lowPublished Aug 27, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		18.2 99% confidence 99 percent, High
93 Qwen3.6 35B-A3B Alibaba / Qwen 17.5%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Aug 29, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		17.5 37% confidence 37 percent, Low
94 Qwen3.6 Flash Alibaba / Qwen 17.2%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Aug 29, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		17.2 32% confidence 32 percent, Low
95 GPT-5.4 mini OpenAI 17.2%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] nonePublished Aug 28, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		17.2 100% confidence 100 percent, Full
96 o4-mini OpenAI 16.1%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] lowPublished Aug 27, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		16.1 87% confidence 87 percent, High
97 Claude Opus 4.1 Anthropic 12.6%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] 32KPublished Jun 11, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		12.6 80% confidence 80 percent, High
98 Qwen3.5 Flash Alibaba / Qwen 9.5%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Aug 28, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		9.5 64% confidence 64 percent, Medium
99 GPT-4.1 mini OpenAI 6.7%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Aug 27, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		6.7 87% confidence 87 percent, High
100 GPT-4.1 OpenAI 6.0%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Aug 27, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		6.0 100% confidence 100 percent, Full
101 GPT-5 Mini OpenAI 6.0%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] minimalPublished Aug 27, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		6.0 99% confidence 99 percent, High
102 GPT-5 Nano OpenAI 6.0%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] lowPublished Aug 27, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		6.0 85% confidence 85 percent, High
103 GPT-5.4 nano OpenAI 4.6%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] nonePublished Aug 28, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		4.6 100% confidence 100 percent, Full
104 GPT-5 Nano OpenAI 1.8%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] minimalPublished Aug 27, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		1.8 85% confidence 85 percent, High
105 GPT-4o mini OpenAI 0.7%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Aug 27, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		0.7 100% confidence 100 percent, Full
106 GPT-4o (2024-08-06) OpenAI 0.4%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Aug 27, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 		0.4 100% confidence 100 percent, Full

Results are as published by the source behind each value (hover or tap the number). Benchmark scores are shown individually for every source/variant (a model may have multiple rows), and vary by version, harness and date; the normalized column uses the method’s fixed 0–100 scales and feeds the math pillar of the SI Score.

What changed

All releases · RSS feed

Alerts on this device

What to be alerted about
RSS feed