LiveCodeBench (Vals AI) leaderboard
- Claude Fable 5.1 leads
- 90.5%
- models measured
- 95
- lead over #2
- +0.7 pts
- median
- 83.2%
Leader on LiveCodeBench (Vals AI) · read October 21
# LiveCodeBench (Vals AI), read October 21
leaders = {
"Claude Fable 5.1": 90.5%, "Claude Fable 5": 89.8%, "Gemini 3.8 Flash": 89.5%,
}
On October 21, 2026, 99 of our catalog's models had a LiveCodeBench (Vals AI) result, and Anthropic Claude Fable 5.1 led with 90.5%, ahead of Anthropic Claude Fable 5 at 89.8%.
Competitive programming problems collected after the models were trained, run by Vals AI. Share solved.
Role in the index
A coding board we show beside the Vibe Coding Index, not inside it: the index keeps the six boards it was defined on, so a score does not move every time a board is added. Read the methodology
| # | Model | Result | Latency, as printed | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|---|
| 1 | Claude Fable 5.1Anthropic | 90.5% | 60 s | 84.2 | $50.00 | |
| 2 | Claude Fable 5Anthropic | 89.8% | 1m 59s | 79.3 | $50.00 | |
| 3 | Gemini 3.8 FlashGoogle | 89.5% | 24 s | 61.6 | $3.75 | |
| 4 | Claude Opus 5Anthropic | 89.0% | 59 s | 83.0 | $25.00 | |
| 5 | Gemini 3.7 FlashGoogle | 88.7% | 15 s | 58.0 | $3.75 | |
| 6 | Gemini 3.1 Pro PreviewGoogle | 88.5% | 1m 29s | 32.3 | $12.00 | |
| 7 | Grok 4.6xAI | 88.2% | 1m 58s | 65.7 | $6.00 | |
| 8 | Gemini 3.6 FlashGoogle | 88.1% | 32 s | 62.5 | $3.75 | |
| 9 | GPT-5.2-CodexOpenAI | 88.0% | 2m 48s | 45.4 | $14.00 | |
| 10 | Qwen3.8 Max (0902)Qwen | 87.8% | 5m 38s | 62.4 | $6.00 | |
| 11 | Claude Opus 4.8Anthropic | 87.8% | 2m 40s | 65.0 | $25.00 | |
| 12 | Gemini 3.5 FlashGoogle | 87.6% | 1m 04s | 39.8 | $9.00 | |
| 13 | DeepSeek V4 Pro 0813DeepSeek | 87.5% | 4m 33s | 56.1 | $1.98 | |
| 14 | DeepSeek V4 Pro 0423DeepSeek | 87.5% | 11m 56s | 38.7 | $2.716 | |
| 15 | Grok 4.5xAI | 87.3% | 9m 49s | 58.6 | $6.00 | |
| 16 | GPT-5.3-CodexOpenAI | 87.3% | 2m 13s | 60.0 | $14.00 | |
| 17 | DeepSeek V4 Flash 0731DeepSeek | 87.3% | 2m 07s | 51.8 | $0.28 | |
| 18 | Kimi K3Moonshot AI | 87.2% | 3m 20s | 62.3 | $15.00 | |
| 19 | Qwen3.7 MaxQwen | 87.1% | 3m 12s | 54.1 | $4.425 | |
| 20 | Kimi K2.6Moonshot AI | 86.8% | 8m 49s | 52.1 | $4.00 | |
| 21 | GPT-5 MiniOpenAI | 86.6% | 34 s | 34.2 | $2.00 | |
| 22 | GPT-5.1OpenAI | 86.5% | 2m 21s | 40.4 | $10.00 | |
| 23 | Nemotron 3 UltraNVIDIA | 86.0% | 2m 41s | 30.0 | $2.40 | |
| 24 | Qwen3.6 PlusQwen | 86.0% | 7m 56s | 44.1 | $1.95 | |
| 25 | GPT-5.6 TerraOpenAI | 85.9% | 44 s | 63.9 | $12.00 | |
| 26 | T Inkling SmallThinking Machines | 85.9% | 5m 02s | 27.3 | $1.20 | |
| 27 | Muse Spark 1.1Meta | 85.9% | 1m 53s | 70.3 | $4.25 | |
| 28 | GPT-5OpenAI | 85.9% | 2m 30s | 38.8 | $10.00 | |
| 29 | Gemini 3 Flash PreviewGoogle | 85.6% | 1m 32s | 43.7 | $3.00 | |
| 30 | GPT-5.1-CodexOpenAI | 85.5% | 3m 54s | 15.3 | $10.00 | |
| 31 | T InklingThinking Machines | 85.5% | 4m 18s | 31.5 | $4.05 | |
| 32 | GPT-5.2OpenAI | 85.4% | 2m 32s | 55.9 | $14.00 | |
| 33 | Qwen3.5 Plus 2026-04-20Qwen | 85.3% | 4m 07s | 39.5 | $1.80 | |
| 34 | GPT-5.5OpenAI | 85.3% | 2m 47s | 71.3 | $30.00 | |
| 35 | Claude Opus 4.7Anthropic | 85.1% | 7m 42s | 67.5 | $25.00 | |
| 36 | Claude Opus 4.6Anthropic | 84.7% | 5m 24s | 57.2 | $25.00 | |
| 37 | Grok 4.3xAI | 84.5% | 54m 09s | 37.4 | $2.50 | |
| 38 | Grok 4.20xAI | 84.3% | 1m 16s | 33.3 | $2.50 | |
| 39 | GPT-5.4OpenAI | 84.1% | 3m 29s | 64.4 | $15.00 | |
| 40 | GPT-5.4 NanoOpenAI | 84.0% | 1m 00s | 44.8 | $1.25 | |
| 41 | Qwen3.8 27BQwen | 84.0% | 2m 29s | 74.2 | $2.55 | |
| 42 | Ling 3.0 FlashinclusionAI | 84.0% | 1m 00s | 29.6 | $0.1215 | |
| 43 | o3OpenAI | 83.9% | 1m 04s | Not benchmarked | $8.00 | |
| 44 | Kimi K2.5Moonshot AI | 83.9% | 3m 20s | 38.6 | $2.85 | |
| 45 | Claude Opus 4.5Anthropic | 83.7% | 2m 44s | 44.0 | $25.00 | |
| 46 | GPT-5.1-Codex-MaxOpenAI | 83.6% | 1m 06s | Not benchmarked | $10.00 | |
| 47 | Qwen3.5-FlashQwen | 83.3% | 3m 29s | 41.1 | $0.26 | |
| 48 | gpt-oss-120bOpenAI | 83.2% | 1m 22s | Not benchmarked | $0.50 | |
| 49 | GPT-5.6 SolOpenAI | 82.6% | 57 s | 73.5 | $20.00 | |
| 50 | Claude Sonnet 5Anthropic | 82.4% | 1m 17s | 57.7 | $10.00 | |
| 51 | GLM 4.7Z.ai | 82.2% | 6m 33s | 55.1 | $2.20 | |
| 52 | o4 MiniOpenAI | 82.2% | 33 s | Not benchmarked | $4.40 | |
| 53 | MiniMax M3MiniMax | 82.2% | 6m 07s | 39.2 | $1.20 | |
| 54 | Claude Sonnet 4.6Anthropic | 82.1% | 4m 55s | 57.3 | $15.00 | |
| 55 | Kimi K2.7 CodeMoonshot AI | 82.0% | 3m 34s | 53.4 | $4.00 | |
| 56 | GLM 5Z.ai | 81.9% | 4m 50s | 41.7 | $3.20 | |
| 57 | MiniMax M2.1MiniMax | 81.8% | 4m 06s | Not benchmarked | $1.20 | |
| 58 | MiMo-V2.5Xiaomi | 81.5% | 1m 49s | 50.0 | $0.336 | |
| 59 | GPT-5.4 MiniOpenAI | 81.5% | 3m 09s | 35.8 | $4.50 | |
| 60 | GLM 5.1Z.ai | 81.4% | 3m 46s | 49.2 | $4.40 | |
| 61 | MiMo-V2.5-ProXiaomi | 81.3% | 4m 31s | 31.6 | $0.9152 | |
| 62 | GLM 4.6Z.ai | 81.0% | 3m 56s | 8.8 | $2.20 | |
| 63 | DeepSeek V3.2DeepSeek | 80.7% | 4m 14s | 30.0 | $0.88 | |
| 64 | GLM 5.3Z.ai | 80.5% | 4m 07s | 70.6 | $4.40 | |
| 65 | GLM 5.3 FlashZ.ai | 80.5% | 5m 02s | 57.7 | $0.50 | |
| 66 | gpt-oss-20bOpenAI | 80.4% | 1m 49s | Not benchmarked | $0.15 | |
| 67 | Gemini 3.1 Flash Lite PreviewGoogle | 80.1% | 31 s | 22.7 | $1.50 | |
| 68 | MiniMax M2.7MiniMax | 79.9% | 5m 19s | 36.2 | $1.20 | |
| 69 | MiniMax M2.5MiniMax | 79.2% | 5m 12s | 37.2 | $1.20 | |
| 70 | Gemini 2.5 Pro Preview 06-05Google | 79.2% | 2m 45s | Not benchmarked | $10.00 | |
| 71 | Gemini 3.5 Flash LiteGoogle | 79.0% | 21 s | 49.0 | $2.50 | |
| 72 | Claude Sonnet 4.5Anthropic | 73.0% | 1m 50s | 39.5 | $15.00 | |
| 73 | o3 MiniOpenAI | 71.5% | 54 s | Not benchmarked | $4.40 | |
| 74 | Gemini 2.5 Flash LiteGoogle | 71.4% | 39 s | Not benchmarked | $0.40 | |
| 75 | Kimi K2 0711Moonshot AI | 70.5% | 1m 07s | Not benchmarked | $2.30 | |
| 76 | R1RetiredDeepSeek | 70.2% | 1m 26s | Not benchmarked | $2.50 | |
| 77 | GPT-5 NanoOpenAI | 70.2% | 2m 16s | Not benchmarked | $0.40 | |
| 78 | GLM 5.2Z.ai | 69.5% | 8m 52s | 61.1 | $4.40 | |
| 79 | GLM 4.5Z.ai | 67.5% | 6m 31s | Not benchmarked | $2.20 | |
| 80 | Claude Opus 4.1Anthropic | 66.5% | 1m 32s | Not benchmarked | $75.00 | |
| 81 | DeepSeek V3 0324DeepSeek | 65.5% | 27 s | Not benchmarked | $1.07 | |
| 82 | Kimi K2 ThinkingMoonshot AI | 63.1% | 19m 11s | Not benchmarked | $2.50 | |
| 83 | GPT-4.1 MiniOpenAI | 58.2% | 28 s | Not benchmarked | $1.60 | |
| 84 | Mistral Large 3 2512Mistral AI | 55.3% | 29 s | Not benchmarked | $1.50 | |
| 85 | GPT-4.1OpenAI | 54.7% | 31 s | Not benchmarked | $8.00 | |
| 86 | Devstral 2 2512Mistral AI | 51.8% | 1m 28s | Not benchmarked | $2.00 | |
| 87 | o1OpenAI | 50.3% | 1m 32s | Not benchmarked | $60.00 | |
| 88 | Llama 4 MaverickMeta Llama | 47.3% | 9.4 s | Not benchmarked | $1.00 | |
| 89 | Mistral Medium 3.1Mistral AI | 44.8% | 12 s | Not benchmarked | $2.00 | |
| 90 | Gemini 2.0 FlashRetiredGoogle | 43.6% | 3.4 s | Not benchmarked | $0.40 | |
| 91 | GPT-4o (2024-11-20)OpenAI | 43.4% | 4.3 s | Not benchmarked | $10.00 | |
| 92 | GPT-4.1 NanoOpenAI | 42.7% | 12 s | Not benchmarked | $0.40 | |
| 93 | Gemini 1.5 ProRetiredGoogle | 41.7% | 4.0 s | Not benchmarked | $5.00 | |
| 94 | Claude Haiku 4.5Anthropic | 41.2% | 1m 22s | 35.6 | $5.00 | |
| 95 | Llama 4 ScoutMeta Llama | 38.5% | 17 s | Not benchmarked | $0.50 | |
| 96 | Mistral LargeMistral AI | 37.1% | 5.3 s | Not benchmarked | $6.00 | |
| 97 | Gemini 1.5 FlashRetiredGoogle | 36.9% | 2.5 s | Not benchmarked | $0.30 | |
| 98 | Command ACohere | 35.1% | 12 s | Not benchmarked | $10.00 | |
| 99 | GPT-4o-miniOpenAI | 26.4% | 7.4 s | Not benchmarked | $0.60 |
Only models in our catalog are listed, and each appears once with its best published run. 136 rows were read from the source on October 21, 2026, and rows that match no catalog model are left out. Every other column is a figure the board publishes for the same run.