FrontierCode leaderboard
- Claude Opus 5.5 leads
- 54.6%
- models measured
- 41
- lead over #2
- +1.2 pts
- median
- 42.4%
Leader on FrontierCode · read October 22
# FrontierCode, read October 22
leaders = {
"Claude Opus 5.5": 54.6%, "Claude Fable 5": 53.5%, "Claude Opus 5": 53.4%,
}
On October 22, 2026, 41 of our catalog's models had a FrontierCode result, and Anthropic Claude Opus 5.5 led with 54.6%, ahead of Anthropic Claude Fable 5 at 53.5%.
Epoch AI collects its results; its benchmarking hub describes the tasks and links the authors.
Role in the index
FrontierCode carries 15% of the Vibe Coding Index. A model missing it is scored on the other benchmarks of that index. Read the methodology
| # | Model | Result | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|
| 1 | Claude Opus 5.5Anthropic | 54.6% | 88.8 | $20.00 | |
| 2 | Claude Fable 5Anthropic | 53.5% | 79.3 | $50.00 | |
| 3 | Claude Opus 5Anthropic | 53.4% | 83.0 | $25.00 | |
| 4 | GPT-6 AstraOpenAI | 53.3% | 85.1 | $50.00 | |
| 5 | Claude Sonnet 5.5Anthropic | 52.1% | 87.3 | $10.00 | |
| 6 | Claude Fable 5.1Anthropic | 50.9% | 84.2 | $50.00 | |
| 7 | GPT-6.1 SolOpenAI | 50.2% | 82.4 | $10.00 | |
| 8 | GPT-6 SolOpenAI | 49.3% | 76.0 | $10.00 | |
| 9 | Grok 4.6xAI | 48.0% | 65.7 | $6.00 | |
| 10 | Grok 4.7xAI | 47.6% | 66.7 | $6.00 | |
| 11 | Gemini 4 ArgonGoogle | 47.5% | 81.0 | $10.00 | |
| 12 | GPT-5.6 SolOpenAI | 47.5% | 73.5 | $20.00 | |
| 13 | Claude Opus 4.8Anthropic | 46.5% | 65.0 | $25.00 | |
| 14 | Claude Haiku 5.5Anthropic | 46.4% | 70.7 | $0.50 | |
| 15 | Kimi K3Moonshot AI | 44.2% | 62.3 | $15.00 | |
| 16 | DeepSeek V4.1 FlashDeepSeek | 43.9% | 62.5 | $0.84 | |
| 17 | Gemini 3.7 FlashGoogle | 43.6% | 58.0 | $3.75 | |
| 18 | GPT-5.5OpenAI | 43.0% | 71.3 | $30.00 | |
| 19 | Claude Sonnet 5Anthropic | 42.7% | 57.7 | $10.00 | |
| 20 | Grok 4.5xAI | 42.4% | 58.6 | $6.00 | |
| 21 | GPT-6 LunaOpenAI | 42.4% | 57.3 | $0.50 | |
| 22 | GPT-5.6 TerraOpenAI | 41.3% | 63.9 | $12.00 | |
| 23 | Gemini 3.8 FlashGoogle | 41.2% | 61.6 | $3.75 | |
| 24 | GLM 5.3Z.ai | 40.1% | 70.6 | $4.40 | |
| 25 | GPT-5.6 LunaOpenAI | 39.8% | 58.0 | $1.20 | |
| 26 | Mistral Large 4Mistral AI | 39.4% | 56.4 | $4.18 | |
| 27 | Claude Opus 4.7Anthropic | 38.5% | 67.5 | $25.00 | |
| 28 | Gemini 3.6 FlashGoogle | 34.4% | 62.5 | $3.75 | |
| 29 | GLM 5.3 FlashZ.ai | 31.8% | 57.7 | $0.50 | |
| 30 | Kimi K2.7 CodeMoonshot AI | 30.1% | 53.4 | $4.00 | |
| 31 | DeepSeek V4 Pro 0813DeepSeek | 28.5% | 56.1 | $1.98 | |
| 32 | GPT-5.4 MiniOpenAI | 27.0% | 35.8 | $4.50 | |
| 33 | Claude Opus 4.6Anthropic | 26.6% | 57.2 | $25.00 | |
| 34 | GLM 5.2Z.ai | 24.5% | 61.1 | $4.40 | |
| 35 | Claude Sonnet 4.6Anthropic | 24.3% | 57.3 | $15.00 | |
| 36 | DeepSeek V4 Flash 0731DeepSeek | 18.8% | 51.8 | $0.396 | |
| 37 | DeepSeek V4 Pro 0423DeepSeek | 17.6% | 38.7 | $2.976 | |
| 38 | MiniMax M3MiniMax | 14.7% | 39.2 | $1.20 | |
| 39 | T InklingThinking Machines | 14.0% | 31.5 | $4.05 | |
| 40 | Nemotron 3 UltraNVIDIA | 13.6% | 30.0 | $2.40 | |
| 41 | Qwen3.7 PlusQwen | 10.2% | 33.2 | $1.28 |
Only models in our catalog are listed, and each appears once with its best published run. 215 rows were read from the source on October 22, 2026, and rows that match no catalog model are left out.