Artificial Analysis SciCode (AA-SciCode)
Data verified
An Artificial Analysis SciCode score.
Top models on AA-SciCode — November 13, 2026
As of November 13, 2026, Claude Fable 5.2 leads the AA-SciCode leaderboard with 67.7% , followed by Claude Opus 5.5 (66.9%) and Claude Fable 5.1 (63.1%).
1Closed
Claude Fable 5.2
Anthropic
Independent run: Artificial Analysis67.7%
Overall 93.65Context 1M
2Closed
Claude Opus 5.5
Anthropic
Independent run: Artificial Analysis66.9%
Overall 86.62Context 1M
3Closed
Claude Fable 5.1
Anthropic
Independent run: Artificial Analysis63.1%
Overall 79.21Context 1M
124 modelsCoding5% of Coding reference weightCurrentUpdated November 13, 2026
Leaderboard (124 models)
Of 124 results: 103 independent, 21 carried from an upstream page.
Ranks on this table count the 103 benchmark-run and independent results. A result where who ran it is not recorded keeps its place by score, without a rank.
- Lab-reported
- 0
- Independent
- 103
- Benchmark-run
- 0
- Carried from upstream
- 21
Who ran it
Showing all 124 results, in score order.
124 of 124 rows
| Rank | Model / configuration | Score | Who ran it | Parameters (B) | Open / closed |
|---|---|---|---|---|---|
| 1 | Claude Fable 5.2AnthropicIndependent run: Artificial Analysis | 67.7% | Independent run: Artificial Analysis | Not reported | Closed |
| 2 | Claude Opus 5.5AnthropicIndependent run: Artificial Analysis | 66.9% | Independent run: Artificial Analysis | Not reported | Closed |
| 3 | Claude Fable 5.1AnthropicIndependent run: Artificial Analysis | 63.1% | Independent run: Artificial Analysis | Not reported | Closed |
| 4 | Muse Wonder FullMetaIndependent run: Artificial Analysis | 62.6% | Independent run: Artificial Analysis | Not reported | Closed |
| 5 | Gemini 4 ArgonGoogleIndependent run: Artificial Analysis | 61.8% | Independent run: Artificial Analysis | Not reported | Closed |
| 6 | Claude Fable 5AnthropicIndependent run: Artificial Analysis | 61.0% | Independent run: Artificial Analysis | Not reported | Closed |
| 7 | Claude Sonnet 5.5AnthropicIndependent run: Artificial Analysis | 61.0% | Independent run: Artificial Analysis | Not reported | Closed |
| 8 | MiMo-V2.6-ProXiaomiIndependent run: Artificial Analysis | 60.9% | Independent run: Artificial Analysis | Not reported | Open |
| 9 | Kimi K3.1 ProMoonshot AIIndependent run: Artificial Analysis | 60.8% | Independent run: Artificial Analysis | Not reported | Open |
| 10 | Muse WonderMetaIndependent run: Artificial Analysis | 60.3% | Independent run: Artificial Analysis | Not reported | Closed |
| 11 | Kimi K3Moonshot AIIndependent run: Artificial Analysis | 59.5% | Independent run: Artificial Analysis | Not reported | Open |
| 12 | GLM-5.3Z.AIIndependent run: Artificial Analysis | 59.0% | Independent run: Artificial Analysis | Not reported | Open |
| 13 | Step 5 PreviewStepFunIndependent run: Artificial Analysis | 58.9% | Independent run: Artificial Analysis | Not reported | Open |
| 14 | Muse Spark 1.1MetaIndependent run: Artificial Analysis | 58.8% | Independent run: Artificial Analysis | Not reported | Closed |
| 15 | Muse Spark 1.3MetaIndependent run: Artificial Analysis | 58.8% | Independent run: Artificial Analysis | Not reported | Closed |
| No rank on this table | Gemini 3.1 ProGoogleCarried from an upstream page | 58.7% | Carried from an upstream page | Not reported | Closed |
| 16 | Muse Spark OpenMetaIndependent run: Artificial Analysis | 58.4% | Independent run: Artificial Analysis | Not reported | Open |
| 17 | GPT-6 SolOpenAIIndependent run: Artificial Analysis | 57.6% | Independent run: Artificial Analysis | Not reported | Closed |
| 18 | Muse Spark 1.2MetaIndependent run: Artificial Analysis | 57.4% | Independent run: Artificial Analysis | Not reported | Closed |
| 19 | Grok 4.7xAIIndependent run: Artificial Analysis | 57.4% | Independent run: Artificial Analysis | Not reported | Closed |
| 20 | Gemini 3.7 FlashGoogleIndependent run: Artificial Analysis | 57.2% | Independent run: Artificial Analysis | Not reported | Closed |
| 21 | GPT-5.6 SolOpenAIIndependent run: Artificial Analysis | 57.1% | Independent run: Artificial Analysis | Not reported | Closed |
| 22 | Gemini 3.8 FlashGoogleIndependent run: Artificial Analysis | 56.6% | Independent run: Artificial Analysis | Not reported | Closed |
| 23 | GPT-6 AstraOpenAIIndependent run: Artificial Analysis | 56.5% | Independent run: Artificial Analysis | Not reported | Closed |
| 24 | Grok 4.6xAIIndependent run: Artificial Analysis | 56.5% | Independent run: Artificial Analysis | Not reported | Closed |
| 25 | Claude Opus 5AnthropicIndependent run: Artificial Analysis | 56.4% | Independent run: Artificial Analysis | Not reported | Closed |
| 26 | GPT-5.5OpenAIIndependent run: Artificial Analysis | 55.8% | Independent run: Artificial Analysis | Not reported | Closed |
| 27 | GPT-6.1 TerraOpenAIIndependent run: Artificial Analysis | 55.1% | Independent run: Artificial Analysis | Not reported | Closed |
| 28 | GPT-5.6 TerraOpenAIIndependent run: Artificial Analysis | 55.0% | Independent run: Artificial Analysis | Not reported | Closed |
| 29 | Claude Haiku 5.5AnthropicIndependent run: Artificial Analysis | 55.0% | Independent run: Artificial Analysis | Not reported | Closed |
| 30 | Grok 4.5xAIIndependent run: Artificial Analysis | 55.0% | Independent run: Artificial Analysis | Not reported | Closed |
| 31 | GPT-6 LunaOpenAIIndependent run: Artificial Analysis | 54.6% | Independent run: Artificial Analysis | Not reported | Closed |
| 32 | Claude Opus 4.8AnthropicIndependent run: Artificial Analysis | 54.4% | Independent run: Artificial Analysis | Not reported | Closed |
| 33 | Claude Sonnet 5AnthropicIndependent run: Artificial Analysis | 54.3% | Independent run: Artificial Analysis | Not reported | Closed |
| 34 | GPT-6.1 SolOpenAIIndependent run: Artificial Analysis | 54.2% | Independent run: Artificial Analysis | Not reported | Closed |
| 35 | Ling 3.1 FlashInclusionAIIndependent run: Artificial Analysis | 54.1% | Independent run: Artificial Analysis | Not reported | Open |
| 36 | Gemini 3.5 FlashGoogleIndependent run: Artificial Analysis | 53.9% | Independent run: Artificial Analysis | Not reported | Closed |
| 37 | GPT-5.6 LunaOpenAIIndependent run: Artificial Analysis | 53.6% | Independent run: Artificial Analysis | Not reported | Closed |
| 38 | Gemini 3.6 FlashGoogleIndependent run: Artificial Analysis | 53.4% | Independent run: Artificial Analysis | Not reported | Closed |
| 39 | MiniMax M3.1 FlashMiniMaxIndependent run: Artificial Analysis | 53.4% | Independent run: Artificial Analysis | Not reported | Closed |
| 40 | Inkling 2 SmallThinking Machines LabIndependent run: Artificial Analysis | 52.3% | Independent run: Artificial Analysis | Not reported | Closed |
| 41 | GPT-5.4 miniOpenAIIndependent run: Artificial Analysis | 52.1% | Independent run: Artificial Analysis | Not reported | Closed |
| No rank on this table | Qwen3.8 Max PreviewAlibabaCarried from an upstream page | 52.1% | Carried from an upstream page | Not reported | Closed |
| 42 | DeepSeek V4.1 Flash 1017DeepSeekIndependent run: Artificial Analysis | 51.9% | Independent run: Artificial Analysis | Not reported | Open |
| No rank on this table | Kimi K2.6Moonshot AICarried from an upstream page | 51.5% | Carried from an upstream page | Not reported | Open |
| 43 | MiMo-V2.6-FlashXiaomiIndependent run: Artificial Analysis | 51.3% | Independent run: Artificial Analysis | Not reported | Open |
| 44 | GLM-5.2Z.AIIndependent run: Artificial Analysis | 51.2% | Independent run: Artificial Analysis | Not reported | Open |
| No rank on this table | DeepSeek V4 Pro 0813DeepSeekCarried from an upstream page | 51.0% | Carried from an upstream page | Not reported | Open |
| 45 | MiMo-V2.5-ProXiaomiIndependent run: Artificial Analysis | 50.6% | Independent run: Artificial Analysis | Not reported | Closed |
| 46 | Qwen3.8-Flash-NextAlibabaIndependent run: Artificial Analysis | 50.6% | Independent run: Artificial Analysis | Not reported | Open |
| No rank on this table | DeepSeek V4 Flash 0731DeepSeekCarried from an upstream page | 50.3% | Carried from an upstream page | Not reported | Open |
| 47 | MiniMax M2.7MiniMaxIndependent run: Artificial Analysis | 50.1% | Independent run: Artificial Analysis | Not reported | Open |
| 48 | Inkling-SmallThinking Machines LabIndependent run: Artificial Analysis | 49.7% | Independent run: Artificial Analysis | Not reported | Open |
| 49 | Qwen3.7 MaxAlibabaIndependent run: Artificial Analysis | 49.5% | Independent run: Artificial Analysis | Not reported | Closed |
| 50 | Hy3TencentIndependent run: Artificial Analysis | 48.6% | Independent run: Artificial Analysis | Not reported | Open |
| No rank on this table | Hy3 PreviewTencentCarried from an upstream page | 48.6% | Carried from an upstream page | Not reported | Open |
| 51 | Beam 1.0 501B-A23BReflection AIIndependent run: Artificial Analysis | 48.5% | Independent run: Artificial Analysis | Not reported | Open |
| 52 | Grok 4.3xAIIndependent run: Artificial Analysis | 48.3% | Independent run: Artificial Analysis | Not reported | Closed |
| 53 | Quasar 438BMultiverse ComputingIndependent run: Artificial Analysis | 48.1% | Independent run: Artificial Analysis | Not reported | Closed |
| No rank on this table | Solar Pro 4UpstageCarried from an upstream page | 48.0% | Carried from an upstream page | Not reported | Closed |
| 54 | Kimi K2.7 CodeMoonshot AIIndependent run: Artificial Analysis | 47.8% | Independent run: Artificial Analysis | Not reported | Open |
| No rank on this table | Nemotron 3.5 Super VL 120B A12BNVIDIACarried from an upstream page | 47.8% | Carried from an upstream page | Not reported | Open |
| No rank on this table | Nemotron 3.5 Super VL 120B A12B NVFP4NVIDIACarried from an upstream page | 47.8% | Carried from an upstream page | Not reported | Open |
| 55 | GPT-5.4 nanoOpenAIIndependent run: Artificial Analysis | 47.2% | Independent run: Artificial Analysis | Not reported | Closed |
| 56 | MiniMax M3MiniMaxIndependent run: Artificial Analysis | 47.1% | Independent run: Artificial Analysis | Not reported | Open |
| 57 | InklingThinking Machines LabIndependent run: Artificial Analysis | 47.0% | Independent run: Artificial Analysis | Not reported | Open |
| 58 | Qwen3.8-27BAlibabaIndependent run: Artificial Analysis | 46.6% | Independent run: Artificial Analysis | Not reported | Open |
| 59 | DeepSeek V4 Pro (High)DeepSeekIndependent run: Artificial Analysis | 46.4% | Independent run: Artificial Analysis | Not reported | Open |
| 60 | Gemini 2.5 ProGoogleIndependent run: Artificial Analysis | 46.3% | Independent run: Artificial Analysis | Not reported | Closed |
| 61 | Qwen3.7 PlusAlibabaIndependent run: Artificial Analysis | 46.1% | Independent run: Artificial Analysis | Not reported | Closed |
| 62 | Gemma 4 31BGoogleIndependent run: Artificial Analysis | 45.5% | Independent run: Artificial Analysis | Not reported | Open |
| 63 | Apodex 1.1ApodexIndependent run: Artificial Analysis | 45.5% | Independent run: Artificial Analysis | Not reported | Closed |
| No rank on this table | Apodex 1.1 MiniApodexCarried from an upstream page | 45.5% | Carried from an upstream page | Not reported | Open |
| 64 | Muse Glimmer 30BMetaIndependent run: Artificial Analysis | 44.9% | Independent run: Artificial Analysis | Not reported | Open |
| 65 | GLM-5.1Z.AIIndependent run: Artificial Analysis | 44.8% | Independent run: Artificial Analysis | Not reported | Open |
| No rank on this table | Qwen3.5 397BAlibabaCarried from an upstream page | 44.8% | Carried from an upstream page | Not reported | Open |
| No rank on this table | Qwen3.5 397B (Reasoning)AlibabaCarried from an upstream page | 44.8% | Carried from an upstream page | Not reported | Open |
| 66 | Ling 3.0 Flash VLInclusionAIIndependent run: Artificial Analysis | 44.2% | Independent run: Artificial Analysis | Not reported | Open |
| 67 | Step 3.7 FlashStepFunIndependent run: Artificial Analysis | 43.9% | Independent run: Artificial Analysis | Not reported | Open |
| 68 | Qwen3.6-27BAlibabaIndependent run: Artificial Analysis | 42.8% | Independent run: Artificial Analysis | Not reported | Open |
| 69 | Ling 3.0 FlashInclusionAIIndependent run: Artificial Analysis | 42.0% | Independent run: Artificial Analysis | Not reported | Open |
| No rank on this table | K-EXAONE 2.0LG AI ResearchCarried from an upstream page | 42.0% | Carried from an upstream page | Not reported | Open |
| 70 | Gemini 3.5 Flash-LiteGoogleIndependent run: Artificial Analysis | 41.3% | Independent run: Artificial Analysis | Not reported | Closed |
| 71 | A.X K2SK TelecomIndependent run: Artificial Analysis | 41.0% | Independent run: Artificial Analysis | Not reported | Open |
| 72 | Trinity-Large-ThinkingArcee AIIndependent run: Artificial Analysis | 40.6% | Independent run: Artificial Analysis | Not reported | Open |
| No rank on this table | Trinity-Large-PreviewArcee AICarried from an upstream page | 40.6% | Carried from an upstream page | Not reported | Open |
| No rank on this table | Nemotron 3 UltraNVIDIACarried from an upstream page | 40.3% | Carried from an upstream page | Not reported | Open |
| 73 | Mistral Medium 3.5 128BMistralIndependent run: Artificial Analysis | 40.2% | Independent run: Artificial Analysis | Not reported | Open |
| 74 | Gemma 4 26B A4BGoogleIndependent run: Artificial Analysis | 40.0% | Independent run: Artificial Analysis | Not reported | Open |
| 75 | Qwen3.5-122B-A10BAlibabaIndependent run: Artificial Analysis | 39.7% | Independent run: Artificial Analysis | Not reported | Open |
| 76 | Nemotron 3.5 Bolt 74B-A9BNVIDIAIndependent run: Artificial Analysis | 39.7% | Independent run: Artificial Analysis | Not reported | Open |
| 77 | DeepSeek V3 0324DeepSeekIndependent run: Artificial Analysis | 39.0% | Independent run: Artificial Analysis | Not reported | Open |
| 78 | GPT-5 miniOpenAIIndependent run: Artificial Analysis | 39.0% | Independent run: Artificial Analysis | Not reported | Closed |
| 79 | GPT-OSS 20BOpenAIIndependent run: Artificial Analysis | 38.9% | Independent run: Artificial Analysis | Not reported | Open |
| 80 | Mistral Small 4MistralIndependent run: Artificial Analysis | 38.8% | Independent run: Artificial Analysis | Not reported | Open |
| No rank on this table | North Mini CodeCohereCarried from an upstream page | 38.8% | Carried from an upstream page | Not reported | Open |
| No rank on this table | Mistral Small 4 (Reasoning)MistralCarried from an upstream page | 38.8% | Carried from an upstream page | Not reported | Open |
| 81 | Mercury 2.5InceptionIndependent run: Artificial Analysis | 38.5% | Independent run: Artificial Analysis | Not reported | Closed |
| 82 | Command A+CohereIndependent run: Artificial Analysis | 38.5% | Independent run: Artificial Analysis | Not reported | Open |
| 83 | Granite 4.2 30BIBMIndependent run: Artificial Analysis | 37.8% | Independent run: Artificial Analysis | Not reported | Open |
| 84 | Mercury 2InceptionIndependent run: Artificial Analysis | 37.7% | Independent run: Artificial Analysis | Not reported | Closed |
| 85 | Qwen3.6-35B-A3BAlibabaIndependent run: Artificial Analysis | 36.6% | Independent run: Artificial Analysis | Not reported | Open |
| 86 | Mistral Large 3MistralIndependent run: Artificial Analysis | 36.6% | Independent run: Artificial Analysis | Not reported | Open |
| 87 | DeepSeek V3DeepSeekIndependent run: Artificial Analysis | 35.8% | Independent run: Artificial Analysis | Not reported | Open |
| 88 | GPT-OSS 120BOpenAIIndependent run: Artificial Analysis | 34.0% | Independent run: Artificial Analysis | Not reported | Open |
| No rank on this table | Nemotron 3.5 Lightning 30B A3B NVFP4NVIDIACarried from an upstream page | 32.1% | Carried from an upstream page | Not reported | Open |
| 89 | Llama 4 MaverickMetaIndependent run: Artificial Analysis | 31.7% | Independent run: Artificial Analysis | Not reported | Open |
| 90 | Granite 4.2 8BIBMIndependent run: Artificial Analysis | 31.5% | Independent run: Artificial Analysis | Not reported | Open |
| 91 | Mistral Small 3.1MistralIndependent run: Artificial Analysis | 27.8% | Independent run: Artificial Analysis | Not reported | Open |
| 92 | MiniCPM5-2BOpenBMBIndependent run: Artificial Analysis | 26.3% | Independent run: Artificial Analysis | Not reported | Open |
| 93 | Solar Pro 3UpstageIndependent run: Artificial Analysis | 25.5% | Independent run: Artificial Analysis | Not reported | Closed |
| 94 | Granite 4.2 3BIBMIndependent run: Artificial Analysis | 25.3% | Independent run: Artificial Analysis | Not reported | Open |
| 95 | Gemma 4 E4BGoogleIndependent run: Artificial Analysis | 24.4% | Independent run: Artificial Analysis | Not reported | Open |
| 96 | Ling 3.0 TinyInclusionAIIndependent run: Artificial Analysis | 24.2% | Independent run: Artificial Analysis | Not reported | Open |
| No rank on this table | Ministral 3 14B (Reasoning)MistralCarried from an upstream page | 23.8% | Carried from an upstream page | Not reported | Open |
| 97 | Ministral 3 14BMistralIndependent run: Artificial Analysis | 23.8% | Independent run: Artificial Analysis | Not reported | Open |
| 98 | Gemma 3 27BGoogleIndependent run: Artificial Analysis | 23.3% | Independent run: Artificial Analysis | Not reported | Open |
| 99 | Celeris-1CelerisIndependent run: Artificial Analysis | 21.6% | Independent run: Artificial Analysis | Not reported | Closed |
| 100 | Llama 4 ScoutMetaIndependent run: Artificial Analysis | 21.3% | Independent run: Artificial Analysis | Not reported | Open |
| No rank on this table | Ministral 3 8B (Reasoning)MistralCarried from an upstream page | 20.7% | Carried from an upstream page | Not reported | Open |
| 101 | Ministral 3 8BMistralIndependent run: Artificial Analysis | 20.7% | Independent run: Artificial Analysis | Not reported | Open |
| No rank on this table | Ministral 3 3B (Reasoning)MistralCarried from an upstream page | 15.3% | Carried from an upstream page | Not reported | Open |
| 102 | Ministral 3 3BMistralIndependent run: Artificial Analysis | 15.3% | Independent run: Artificial Analysis | Not reported | Open |
| 103 | LFM2.5-2.6BLiquidAIIndependent run: Artificial Analysis | 14.4% | Independent run: Artificial Analysis | Not reported | Open |
A dash in the rank column marks a result carried from an upstream page, where who ran it is not recorded. It keeps its place by score, without a rank on this table.
- Lab-reported
- Benchmark-run
- Independent
According to BenchLM.ai, Claude Fable 5.2 leads the AA-SciCode benchmark with a score of 67.7%, followed by Claude Opus 5.5 (66.9%) and Claude Fable 5.1 (63.1%). The scores show moderate spread, with meaningful differences between the top tier and mid-tier models.
124 models have been evaluated on AA-SciCode. The benchmark falls in the Coding category. BenchAlign v5.8 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.
About AA-SciCode
Year
2026
Tasks
Scientific coding subproblems
Format
Task success rate
Difficulty
Scientific programming
BenchLM stores the Artificial Analysis SciCode result separately from the core SciCode lane because Artificial Analysis controls the evaluation configuration.
Freshness and provenance
Version
AA-SciCode 2026
Refresh cadence
Quarterly
Staleness state
Current
Question availability
Public benchmark set
Current
BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.
Questions
What does AA-SciCode measure?
An Artificial Analysis SciCode score.
Which model scores highest on AA-SciCode?
Claude Fable 5.2 by Anthropic currently leads with a score of 67.7% on AA-SciCode.
How many models are evaluated on AA-SciCode?
124 AI models have published results on AA-SciCode in the BenchLM catalog.
Last updated: November 13, 2026 · BenchLM version AA-SciCode 2026
Know when it’s worth switching models
The model to choose, the cheaper alternative, and the release we would wait on.
Read a sample issueJoin 7,000+ subscribers.
One email each week. Unsubscribe anytime.