FrontierMath, tier 4 leaderboard
- GPT-6.1 Sol leads
- 100.0%
- models measured
- 64
- lead over #2
- +2.4 pts
- median
- 28.0%
Leader on FrontierMath, tier 4 · read October 21
On October 21, 2026, 64 of our catalog's models had a FrontierMath, tier 4 result, and OpenAI GPT-6.1 Sol led with 100.0%.
The hardest FrontierMath tier, problems that take specialists days. Share solved.
Role in the index
We collect this benchmark and show it as supporting evidence. It is not part of the Vibe Coding Index, because it measures something other than coding and the index is about coding. Read the methodology
| # | Model | Result | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|
| 1 | GPT-6.1 SolOpenAI | 100.0% | 88.7 | $10.00 | |
| 2 | GPT-6 AstraOpenAI | 97.6% | 88.7 | $50.00 | |
| 3 | Claude Opus 5.5Anthropic | 95.0% | 92.2 | $20.00 | |
| 4 | Claude Fable 5Anthropic | 90.2% | 82.1 | $50.00 | |
| 5 | GPT-6 SolOpenAI | 90.0% | 82.8 | $10.00 | |
| 6 | Claude Fable 5.1Anthropic | 87.8% | 89.1 | $50.00 | |
| 7 | GPT-5.6 SolOpenAI | 82.9% | 76.7 | $20.00 | |
| 8 | Claude Sonnet 5.5Anthropic | 80.5% | 88.9 | $10.00 | |
| 9 | GPT-5.6 Sol ProOpenAI | 80.5% | Not benchmarked | $20.00 | |
| 10 | Gemini 4 ArgonGoogle | 78.0% | 80.5 | $10.00 | |
| 11 | GPT-5.5 ProOpenAI | 78.0% | Not benchmarked | $180.00 | |
| 12 | Claude Opus 5Anthropic | 73.2% | 81.8 | $25.00 | |
| 13 | GPT-5.5OpenAI | 72.5% | 67.8 | $30.00 | |
| 14 | GPT-5.6 TerraOpenAI | 70.7% | 69.0 | $12.00 | |
| 15 | GPT-5.6 LunaOpenAI | 61.0% | 61.5 | $1.20 | |
| 16 | GPT-5.4 ProOpenAI | 58.5% | Not benchmarked | $180.00 | |
| 17 | Claude Opus 4.8Anthropic | 56.1% | 66.0 | $25.00 | |
| 18 | GPT-6 LunaOpenAI | 56.1% | 71.4 | $0.50 | |
| 19 | GPT-5.4OpenAI | 49.0% | 37.7 | $15.00 | |
| 20 | Claude Haiku 5.5Anthropic | 46.3% | 75.9 | $0.50 | |
| 21 | Muse Spark 1.3Meta | 46.3% | 79.7 | $4.25 | |
| 22 | Qwen3.8 Max (0902)Qwen | 46.3% | 63.5 | $6.00 | |
| 23 | GPT-5.2 ProOpenAI | 46.0% | Not benchmarked | $168.00 | |
| 24 | Kimi K3Moonshot AI | 39.0% | 61.8 | $11.936 | |
| 25 | Gemini 3.7 FlashGoogle | 36.6% | 69.9 | $3.75 | |
| 26 | Qwen3.7 MaxQwen | 34.1% | 45.1 | $4.425 | |
| 27 | Claude Opus 4.7Anthropic | 31.7% | 51.1 | $25.00 | |
| 28 | Grok 4.6xAI | 31.7% | 76.4 | $6.00 | |
| 29 | GPT-5.2OpenAI | 31.7% | 58.9 | $14.00 | |
| 30 | Claude Sonnet 5Anthropic | 29.3% | 63.5 | $10.00 | |
| 31 | GLM 5.2Z.ai | 29.3% | 57.7 | $7.00 | |
| 32 | GLM 5.3Z.ai | 29.3% | 73.1 | $1.40 | |
| 33 | Claude Opus 4.6Anthropic | 26.8% | 66.9 | $25.00 | |
| 34 | DeepSeek V4 Pro 0813DeepSeek | 26.8% | 54.6 | $1.98 | |
| 35 | DeepSeek V4.1 FlashDeepSeek | 26.8% | 77.2 | $0.525 | |
| 36 | Gemini 3.1 Pro PreviewGoogle | 26.8% | 35.3 | $12.00 | |
| 37 | Gemini 3.5 FlashGoogle | 26.8% | 42.8 | $9.00 | |
| 38 | Mistral Large 4Mistral AI | 26.8% | 55.3 | $4.18 | |
| 39 | Kimi K2.6Moonshot AI | 25.6% | 44.2 | $3.41 | |
| 40 | DeepSeek V4 Flash 0731DeepSeek | 24.4% | Not benchmarked | $1.32 | |
| 41 | Grok 4.5xAI | 24.4% | 75.5 | $6.00 | |
| 42 | Gemini 3.6 FlashGoogle | 22.0% | 48.3 | $3.75 | |
| 43 | Gemini 3.8 FlashGoogle | 22.0% | 70.5 | $4.125 | |
| 44 | GPT-5OpenAI | 22.0% | 72.2 | $10.00 | |
| 45 | GPT-5 ProOpenAI | 19.5% | Not benchmarked | $120.00 | |
| 46 | Gemini 3 Flash PreviewGoogle | 17.1% | 75.8 | $3.00 | |
| 47 | Grok 4.7xAI | 17.1% | 75.0 | $6.00 | |
| 48 | T Inkling SmallThinking Machines | 17.1% | 29.9 | $1.20 | |
| 49 | GLM 5.3 FlashZ.ai | 17.1% | 71.0 | $0.50 | |
| 50 | Grok 4.3xAI | 14.6% | 22.5 | $2.50 | |
| 51 | Kimi K2.7 CodeMoonshot AI | 12.2% | 39.1 | $3.35 | |
| 52 | GPT-5 MiniOpenAI | 12.2% | 59.8 | $2.00 | |
| 53 | GPT-5.4 NanoOpenAI | 12.2% | Not benchmarked | $1.25 | |
| 54 | GPT-5.4 MiniOpenAI | 9.8% | 28.2 | $4.50 | |
| 55 | Claude Opus 4.5Anthropic | 4.9% | 66.8 | $25.00 | |
| 56 | o4 MiniOpenAI | 4.9% | 68.6 | $4.40 | |
| 57 | T InklingThinking Machines | 4.9% | 49.7 | $4.05 | |
| 58 | Claude Opus 4.1Anthropic | 2.4% | 27.2 | $75.00 | |
| 59 | Claude Sonnet 4.5Anthropic | 2.4% | 59.1 | $15.00 | |
| 60 | DeepSeek V4 Pro 0423DeepSeek | 2.4% | 35.3 | $1.9105 | |
| 61 | GPT-5 NanoOpenAI | 2.4% | 34.8 | $0.40 | |
| 62 | Gemini 2.5 ProGoogle | 0.0% | 37.0 | $10.00 | |
| 63 | Gemini 3.5 Flash LiteGoogle | 0.0% | 33.6 | $2.50 | |
| 64 | o3 MiniOpenAI | 0.0% | 52.1 | $4.40 |
Only models in our catalog are listed, and each appears once with its best published run. 197 rows were read from the source on October 21, 2026, and rows that match no catalog model are left out.