SciCode leaderboard
- Claude Opus 5.5 leads
- 66.9%
- models measured
- 108
- lead over #2
- +3.8 pts
- median
- 48.4%
Leader on SciCode · read October 22
# SciCode, read October 22
leaders = {
"Claude Opus 5.5": 66.9%, "Claude Fable 5.1": 63.1%, "Gemini 4 Argon": 61.8%,
}
On October 22, 2026, 108 of our catalog's models had a SciCode result, and Anthropic Claude Opus 5.5 led with 66.9%, ahead of Anthropic Claude Fable 5.1 at 63.1%.
Scientific computing problems written by researchers, solved in code. Share solved. Run by Artificial Analysis on its own harness.
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 | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|
| 1 | Claude Opus 5.5Anthropic | 66.9% | 88.8 | $20.00 | |
| 2 | Claude Fable 5.1Anthropic | 63.1% | 84.2 | $50.00 | |
| 3 | Gemini 4 ArgonGoogle | 61.8% | 81.0 | $10.00 | |
| 4 | Claude Fable 5Anthropic | 61.0% | 79.3 | $50.00 | |
| 5 | Claude Sonnet 5.5Anthropic | 61.0% | 87.3 | $10.00 | |
| 6 | MiMo-V2.6-ProXiaomi | 60.9% | 65.4 | $0.87 | |
| 7 | Muse WonderMeta | 60.3% | 71.9 | $20.00 | |
| 8 | Gemini 3.7 FlashGoogle | 59.8% | 58.0 | $3.75 | |
| 9 | Muse Spark 1.3Meta | 59.7% | 57.4 | $4.25 | |
| 10 | Kimi K3Moonshot AI | 59.5% | 62.3 | $15.00 | |
| 11 | GLM 5.3Z.ai | 59.0% | 70.6 | $4.40 | |
| 12 | Step 5 PreviewStepFun | 58.9% | 61.1 | $2.70 | |
| 13 | Muse Spark 1.1Meta | 58.8% | 70.3 | $4.25 | |
| 14 | Gemini 3.1 Pro PreviewGoogle | 58.7% | 32.3 | $12.00 | |
| 15 | Muse Spark OpenMeta | 58.4% | 47.4 | $4.25 | |
| 16 | GPT-5.6 SolOpenAI | 57.8% | 73.5 | $20.00 | |
| 17 | Grok 4.7xAI | 57.8% | 66.7 | $6.00 | |
| 18 | GPT-6 SolOpenAI | 57.6% | 76.0 | $10.00 | |
| 19 | Muse Spark 1.2Meta | 57.4% | 55.7 | $4.25 | |
| 20 | Gemini 3.8 FlashGoogle | 56.6% | 61.6 | $3.75 | |
| 21 | GPT-6 AstraOpenAI | 56.5% | 85.1 | $50.00 | |
| 22 | Grok 4.6xAI | 56.5% | 65.7 | $6.00 | |
| 23 | Claude Opus 5Anthropic | 56.4% | 83.0 | $25.00 | |
| 24 | GPT-5.5OpenAI | 56.1% | 71.3 | $30.00 | |
| 25 | DeepSeek V4.1 Flash 1017DeepSeek | 55.9% | 70.3 | $0.60 | |
| 26 | GPT-6.1 SolOpenAI | 55.8% | 82.4 | $10.00 | |
| 27 | Claude Haiku 5.5Anthropic | 55.0% | 70.7 | $0.50 | |
| 28 | GPT-5.6 TerraOpenAI | 55.0% | 63.9 | $12.00 | |
| 29 | Grok 4.5xAI | 55.0% | 58.6 | $6.00 | |
| 30 | GPT-6 LunaOpenAI | 54.6% | 57.3 | $0.50 | |
| 31 | Claude Opus 4.8Anthropic | 54.4% | 65.0 | $25.00 | |
| 32 | Claude Sonnet 5Anthropic | 54.3% | 57.7 | $10.00 | |
| 33 | Mistral Large 4Mistral AI | 54.2% | 56.4 | $4.18 | |
| 34 | Ling 3.1 FlashinclusionAI | 54.1% | Not benchmarked | $0.225 | |
| 35 | Qwen3.8 2.4T A95BQwen | 54.1% | Not benchmarked | $6.00 | |
| 36 | Gemini 3.5 FlashGoogle | 53.9% | 39.8 | $9.00 | |
| 37 | GPT-5.6 LunaOpenAI | 53.6% | 58.0 | $1.20 | |
| 38 | Gemini 3.6 FlashGoogle | 53.4% | 62.5 | $3.75 | |
| 39 | Qwen3.8 Max (0902)Qwen | 53.2% | 62.4 | $6.00 | |
| 40 | GPT-5.4 MiniOpenAI | 52.1% | 35.8 | $4.50 | |
| 41 | DeepSeek V4.1 FlashDeepSeek | 51.9% | 62.5 | $0.84 | |
| 42 | GLM 5.3 FlashZ.ai | 51.6% | 57.7 | $0.50 | |
| 43 | Kimi K2.6Moonshot AI | 51.5% | 52.1 | $4.00 | |
| 44 | MiMo-V2.6-FlashXiaomi | 51.3% | 54.7 | $0.28 | |
| 45 | GLM 5.2Z.ai | 51.2% | 61.1 | $4.40 | |
| 46 | DeepSeek V4 Pro 0813DeepSeek | 51.0% | 56.1 | $1.98 | |
| 47 | MiMo-V2.5-ProXiaomi | 50.6% | 31.6 | $0.9152 | |
| 48 | DeepSeek V4 Flash 0731DeepSeek | 50.3% | 51.8 | $0.396 | |
| 49 | Claude Sonnet 4.6Anthropic | 50.1% | 57.3 | $15.00 | |
| 50 | MiniMax M2.7MiniMax | 50.1% | 36.2 | $1.20 | |
| 51 | DeepSeek V4 Flash Vision ExpDeepSeek | 49.7% | Not benchmarked | $1.32 | |
| 52 | T Inkling SmallThinking Machines | 49.7% | 27.3 | $1.20 | |
| 53 | Qwen3.7 MaxQwen | 49.5% | 54.1 | $4.425 | |
| 54 | Hy3Tencent | 48.6% | Not benchmarked | $0.528 | |
| 55 | Grok 4.3xAI | 48.3% | 37.4 | $2.50 | |
| 56 | Solar Pro 4Upstage | 48.0% | Not benchmarked | $0.60 | |
| 57 | Kimi K2.7 CodeMoonshot AI | 47.8% | 53.4 | $4.00 | |
| 58 | Nemotron 3.5 Super VLNVIDIA | 47.8% | Not benchmarked | $0.50 | |
| 59 | Solar Mini 4Upstage | 47.6% | Not benchmarked | $0.20 | |
| 60 | GPT-5.4 NanoOpenAI | 47.2% | 44.8 | $1.25 | |
| 61 | MiniMax M3MiniMax | 47.1% | 39.2 | $1.20 | |
| 62 | T InklingThinking Machines | 47.0% | 31.5 | $4.05 | |
| 63 | Qwen3.8 27BQwen | 46.6% | 74.2 | $2.55 | |
| 64 | Gemini 2.5 ProGoogle | 46.3% | 19.1 | $10.00 | |
| 65 | Qwen3.7 PlusQwen | 46.1% | 33.2 | $1.28 | |
| 66 | Gemma 4 31BGoogle | 45.5% | 39.0 | $0.40 | |
| 67 | Qwen3.5 397B A17BQwen | 44.8% | Not benchmarked | $2.34 | |
| 68 | GLM 5.1Z.ai | 44.8% | 49.2 | $4.40 | |
| 69 | Ling 3.0 Flash VLinclusionAI | 44.2% | Not benchmarked | $0.18 | |
| 70 | Step 3.7 FlashStepFun | 43.9% | Not benchmarked | $1.15 | |
| 71 | MiMo-V2.5Xiaomi | 43.9% | 50.0 | $0.336 | |
| 72 | Gemini 3.1 Flash LiteGoogle | 43.4% | Not benchmarked | $1.50 | |
| 73 | o3 MiniOpenAI | 42.8% | Not benchmarked | $4.40 | |
| 74 | Qwen3.6 27BQwen | 42.8% | 36.8 | $2.70 | |
| 75 | Ling 3.0 FlashinclusionAI | 42.0% | 29.6 | $0.1215 | |
| 76 | Qwen3 235B A22B Instruct 2507Qwen | 41.4% | Not benchmarked | $0.75 | |
| 77 | Gemini 3.5 Flash LiteGoogle | 41.3% | 49.0 | $2.50 | |
| 78 | Trinity Large ThinkingArcee AI | 40.6% | Not benchmarked | $0.80 | |
| 79 | Nemotron 3 UltraNVIDIA | 40.3% | 30.0 | $2.40 | |
| 80 | Mistral Medium 3.5Mistral AI | 40.2% | 25.5 | $7.50 | |
| 81 | Gemma 4 26B A4BGoogle | 40.0% | Not benchmarked | $0.60 | |
| 82 | Qwen3.5-122B-A10BQwen | 39.7% | Not benchmarked | $2.08 | |
| 83 | DeepSeek V3 0324DeepSeek | 39.0% | Not benchmarked | $1.07 | |
| 84 | GPT-5 MiniOpenAI | 39.0% | 34.2 | $2.00 | |
| 85 | gpt-oss-20bOpenAI | 38.9% | Not benchmarked | $0.15 | |
| 86 | Mistral Small 4Mistral AI | 38.8% | Not benchmarked | $0.60 | |
| 87 | Command ACohere | 38.5% | Not benchmarked | $10.00 | |
| 88 | Mercury 2.5Inception | 38.5% | 1.6 | $0.15 | |
| 89 | DeepSeek V3.1 TerminusDeepSeek | 38.0% | Not benchmarked | $1.0142 | |
| 90 | Mercury 2Inception | 37.7% | Not benchmarked | $0.75 | |
| 91 | Qwen3.6 35B A3BQwen | 36.6% | Not benchmarked | $1.00 | |
| 92 | LongCat 2.0Meituan | 36.3% | Not benchmarked | $1.20 | |
| 93 | Nemotron 3 SuperNVIDIA | 36.2% | Not benchmarked | $0.45 | |
| 94 | Qwen3 Coder NextQwen | 36.2% | Not benchmarked | $0.80 | |
| 95 | Qwen3 32BQwen | 36.0% | Not benchmarked | $0.40 | |
| 96 | gpt-oss-120bOpenAI | 34.0% | Not benchmarked | $0.50 | |
| 97 | Qwen3 30B A3B Instruct 2507Qwen | 33.0% | Not benchmarked | $0.30 | |
| 98 | Mistral Medium 3.1Mistral AI | 32.4% | Not benchmarked | $2.00 | |
| 99 | Nemotron 3.5 LightningNVIDIA | 32.1% | 22.6 | $0.147 | |
| 100 | Llama 4 MaverickMeta Llama | 31.7% | Not benchmarked | $1.00 | |
| 101 | Granite 4.2 8BIBM Granite | 31.5% | Not benchmarked | $0.20 | |
| 102 | Qwen3 14BQwen | 30.7% | Not benchmarked | $0.23 | |
| 103 | Nemotron 3 Nano 30B A3BNVIDIA | 30.6% | Not benchmarked | $0.22 | |
| 104 | Qwen3.5-9BQwen | 29.5% | Not benchmarked | $0.15 | |
| 105 | Solar Pro 3Upstage | 25.5% | Not benchmarked | $0.60 | |
| 106 | Gemma 3 27BGoogle | 23.3% | Not benchmarked | $0.30 | |
| 107 | Llama 4 ScoutMeta Llama | 21.3% | Not benchmarked | $0.50 | |
| 108 | Gemma 3 12BGoogle | 16.4% | Not benchmarked | $0.1625 |
Only models in our catalog are listed, and each appears once with its best published run. 199 rows were read from the source on October 22, 2026, and rows that match no catalog model are left out.