# Artificial Analysis SciCode (AA-SciCode)

> An Artificial Analysis SciCode score.

Canonical page: https://benchlm.ai/benchmarks/aascicode

- Category: [Coding](/coding)
- Last updated: September 27, 2026

## About AA-SciCode

- Year: 2026
- Tasks: Scientific coding subproblems
- Format: Task success rate
- Difficulty: Scientific programming
- Paper: [Artificial Analysis SciCode Benchmark Leaderboard](https://artificialanalysis.ai/evaluations/scicode)

BenchLM stores the Artificial Analysis SciCode result separately from the core SciCode lane because Artificial Analysis controls the evaluation configuration.

BenchAlign v5.7 gives AA-SciCode 5% of the Coding reference weight, so it moves the Coding leaderboard and the overall ranking. Reference weights are relative weights in the calibrated model, not fixed shares of a score.

## Leaderboard (94 models)

| Rank | Model | Creator | Score |
|------|-------|---------|-------|
| 1 | [Claude Opus 5.5](/models/claude-opus-5-5) | Anthropic | 66.9% |
| 2 | [Claude Fable 5.1](/models/claude-fable-5-1) | Anthropic | 63.1% |
| 3 | [Claude Fable 5](/models/claude-fable) | Anthropic | 61.0% |
| 4 | [MiMo-V2.6-Pro](/models/mimo-v2-6-pro) | Xiaomi | 60.9% |
| 5 | [Kimi K3](/models/kimi-k3) | Moonshot AI | 59.5% |
| 6 | [GLM-5.3](/models/glm-5-3) | Z.AI | 59.0% |
| 7 | [Step 5 Preview](/models/step-5-preview) | StepFun | 58.9% |
| 8 | [Muse Spark 1.1](/models/muse-spark-1-1) | Meta | 58.8% |
| 9 | [Muse Spark 1.3](/models/muse-spark-1-3) | Meta | 58.8% |
| 10 | [Gemini 3.1 Pro](/models/gemini-3-1-pro) | Google | 58.7% |
| 11 | [GPT-6 Sol](/models/gpt-6-sol) | OpenAI | 57.6% |
| 12 | [Muse Spark 1.2](/models/muse-spark-1-2) | Meta | 57.4% |
| 13 | [Grok 4.7](/models/grok-4-7) | xAI | 57.4% |
| 14 | [Gemini 3.7 Flash](/models/gemini-3-7-flash) | Google | 57.2% |
| 15 | [GPT-5.6 Sol](/models/gpt-5-6-sol) | OpenAI | 57.1% |
| 16 | [Gemini 3.8 Flash](/models/gemini-3-8-flash) | Google | 56.6% |
| 17 | [GPT-6 Astra](/models/gpt-6-astra) | OpenAI | 56.5% |
| 18 | [Grok 4.6](/models/grok-4-6) | xAI | 56.5% |
| 19 | [Claude Opus 5](/models/claude-opus-5) | Anthropic | 56.4% |
| 20 | [GPT-5.5](/models/gpt-5-5) | OpenAI | 55.8% |
| 21 | [GPT-5.6 Terra](/models/gpt-5-6-terra) | OpenAI | 55.0% |
| 22 | [Grok 4.5](/models/grok-4-5) | xAI | 55.0% |
| 23 | [GPT-6 Luna](/models/gpt-6-luna) | OpenAI | 54.6% |
| 24 | [Claude Opus 4.8](/models/claude-opus-4-8) | Anthropic | 54.4% |
| 25 | [Claude Sonnet 5](/models/claude-sonnet-5) | Anthropic | 54.3% |
| 26 | [Gemini 3.5 Flash](/models/gemini-3-5-flash) | Google | 53.9% |
| 27 | [GPT-5.6 Luna](/models/gpt-5-6-luna) | OpenAI | 53.6% |
| 28 | [Gemini 3.6 Flash](/models/gemini-3-6-flash) | Google | 53.4% |
| 29 | [GPT-5.4 mini](/models/gpt-5-4-mini) | OpenAI | 52.1% |
| 30 | [Qwen3.8 Max Preview](/models/qwen3-8-max-preview) | Alibaba | 52.1% |
| 31 | [DeepSeek V4.1 Flash](/models/deepseek-v4-1-flash) | DeepSeek | 51.9% |
| 32 | [Kimi K2.6](/models/kimi-2-6) | Moonshot AI | 51.5% |
| 33 | [GLM-5.2](/models/glm-5-2) | Z.AI | 51.2% |
| 34 | [DeepSeek V4 Pro 0813](/models/deepseek-v4-pro-0813) | DeepSeek | 51.0% |
| 35 | [Qwen3.8-Flash-Next](/models/qwen3-8-flash-next) | Alibaba | 50.6% |
| 36 | [MiMo-V2.5-Pro](/models/mimo-v2-5-pro) | Xiaomi | 50.6% |
| 37 | [DeepSeek V4 Flash 0731](/models/deepseek-v4-flash-0731) | DeepSeek | 50.3% |
| 38 | [MiniMax M2.7](/models/minimax-m2-7) | MiniMax | 50.1% |
| 39 | [Inkling-Small](/models/inkling-small) | Thinking Machines Lab | 49.7% |
| 40 | [Qwen3.7 Max](/models/qwen3-7-max) | Alibaba | 49.5% |
| 41 | [Hy3](/models/hy3) | Tencent | 48.6% |
| 42 | [Hy3 Preview](/models/hy3-preview) | Tencent | 48.6% |
| 43 | [Grok 4.3](/models/grok-4-3) | xAI | 48.3% |
| 44 | [Quasar 438B](/models/quasar-438b) | Multiverse Computing | 48.1% |
| 45 | [Kimi K2.7 Code](/models/kimi-k2-7-code) | Moonshot AI | 47.8% |
| 46 | [GPT-5.4 nano](/models/gpt-5-4-nano) | OpenAI | 47.2% |
| 47 | [MiniMax M3](/models/minimax-m3) | MiniMax | 47.1% |
| 48 | [Inkling](/models/inkling) | Thinking Machines Lab | 47.0% |
| 49 | [Qwen3.8-27B](/models/qwen3-8-27b) | Alibaba | 46.6% |
| 50 | [Gemini 2.5 Pro](/models/gemini-2-5-pro) | Google | 46.3% |
| 51 | [Qwen3.7 Plus](/models/qwen3-7-plus) | Alibaba | 46.1% |
| 52 | [Apodex 1.1](/models/apodex-1-1) | Apodex | 45.5% |
| 53 | [Gemma 4 31B](/models/gemma-4-31b) | Google | 45.5% |
| 54 | [Apodex 1.1 Mini](/models/apodex-1-1-mini) | Apodex | 45.5% |
| 55 | [Muse Glimmer 30B](/models/muse-glimmer-30b) | Meta | 44.9% |
| 56 | [GLM-5.1](/models/glm-5-1) | Z.AI | 44.8% |
| 57 | [Solar Pro 4](/models/solar-pro-4) | Upstage | 44.6% |
| 58 | [Ling 3.0 Flash VL](/models/ling-3-0-flash-vl) | InclusionAI | 44.2% |
| 59 | [Step 3.7 Flash](/models/step-3-7-flash) | StepFun | 43.9% |
| 60 | [Qwen3.6-27B](/models/qwen3-6-27b) | Alibaba | 42.8% |
| 61 | [Ling 3.0 Flash](/models/ling-3-0-flash) | InclusionAI | 42.0% |
| 62 | [Ling 3.0 Flash FP8](/models/ling-3-0-flash-fp8) | InclusionAI | 42.0% |
| 63 | [K-EXAONE 2.0](/models/k-exaone-2-0) | LG AI Research | 42.0% |
| 64 | [Gemini 3.5 Flash-Lite](/models/gemini-3-5-flash-lite) | Google | 41.3% |
| 65 | [A.X K2](/models/a-x-k2) | SK Telecom | 41.0% |
| 66 | [Trinity-Large-Thinking](/models/trinity-large-thinking) | Arcee AI | 40.6% |
| 67 | [Trinity-Large-Preview](/models/trinity-large-preview) | Arcee AI | 40.6% |
| 68 | [Nemotron 3 Ultra](/models/nemotron-3-ultra) | NVIDIA | 40.3% |
| 69 | [Mistral Medium 3.5 128B](/models/mistral-medium-3-5-128b) | Mistral | 40.2% |
| 70 | [Gemma 4 26B A4B](/models/gemma-4-26b-a4b) | Google | 40.0% |
| 71 | [Qwen3.5-122B-A10B](/models/qwen3-5-122b-a10b) | Alibaba | 39.7% |
| 72 | [GPT-OSS 20B](/models/gpt-oss-20b) | OpenAI | 38.9% |
| 73 | [Mistral Small 4](/models/mistral-small-4) | Mistral | 38.8% |
| 74 | [North Mini Code](/models/north-mini-code-1-0) | Cohere | 38.8% |
| 75 | [Mistral Small 4 (Reasoning)](/models/mistral-small-4-reasoning) | Mistral | 38.8% |
| 76 | [Command A+](/models/command-a-plus) | Cohere | 38.5% |
| 77 | [Granite 4.2 30B](/models/granite-4-2-30b) | IBM | 37.8% |
| 78 | [Qwen3.6-35B-A3B](/models/qwen3-6-35b-a3b) | Alibaba | 36.6% |
| 79 | [Mistral Large 3](/models/mistral-large-3) | Mistral | 36.6% |
| 80 | [Nemotron 3 Super 100B](/models/nemotron-3-super-100b) | NVIDIA | 36.2% |
| 81 | [DeepSeek V3](/models/deepseek-v3) | DeepSeek | 35.8% |
| 82 | [GPT-OSS 120B](/models/gpt-oss-120b) | OpenAI | 34.0% |
| 83 | [Nemotron 3.5 Lightning 30B A3B NVFP4](/models/nemotron-3-5-lightning-30b-a3b-nvfp4) | NVIDIA | 32.1% |
| 84 | [Llama 4 Maverick](/models/llama-4-maverick) | Meta | 31.7% |
| 85 | [Granite 4.2 8B](/models/granite-4-2-8b) | IBM | 31.5% |
| 86 | [Nemotron 3 Nano 30B](/models/nemotron-3-nano-30b) | NVIDIA | 30.6% |
| 87 | [MiniCPM5-2B](/models/minicpm5-2b) | OpenBMB | 26.3% |
| 88 | [Solar Pro 3](/models/solar-pro-3) | Upstage | 25.5% |
| 89 | [Granite 4.2 3B](/models/granite-4-2-3b) | IBM | 25.3% |
| 90 | [Ling 3.0 Tiny](/models/ling-3-0-tiny) | InclusionAI | 24.2% |
| 91 | [Gemma 3 27B](/models/gemma-3-27b) | Google | 23.3% |
| 92 | [Celeris-1](/models/celeris-1) | Celeris | 21.6% |
| 93 | [Llama 4 Scout](/models/llama-4-scout) | Meta | 21.3% |
| 94 | [LFM2.5-2.6B](/models/lfm2-5-2-6b) | LiquidAI | 14.4% |

## FAQ

### What does AA-SciCode measure?

An Artificial Analysis SciCode score.

### Which model scores highest on AA-SciCode?

Claude Opus 5.5 by Anthropic currently leads with a score of 66.9% on AA-SciCode.

### How many models are evaluated on AA-SciCode?

94 AI models have been evaluated on AA-SciCode on BenchLM.

### Does AA-SciCode affect BenchLM's overall score?

Yes. BenchAlign v5.7 gives AA-SciCode 5% of the Coding reference weight, so it moves the Coding leaderboard and the overall ranking. Reference weights are relative weights in the calibrated model, not fixed shares of a score.

## Compare Top Models on AA-SciCode

- [Claude Opus 5.5 vs Claude Fable 5.1](/compare/claude-fable-5-1-vs-claude-opus-5-5)
- [Claude Fable 5.1 vs Claude Fable 5](/compare/claude-fable-vs-claude-fable-5-1)
- [Claude Fable 5 vs MiMo-V2.6-Pro](/compare/claude-fable-vs-mimo-v2-6-pro)
- [MiMo-V2.6-Pro vs Kimi K3](/compare/kimi-k3-vs-mimo-v2-6-pro)