Benchmark profile

Artificial Analysis Coding Index (AA Coding Index)

A display-only Artificial Analysis coding index.

Data verified July 29, 2026

Benchmark score on AA Coding Index — July 29, 2026

BenchLM mirrors the published score view for AA Coding Index. Claude Opus 5 leads the public snapshot at 78.0% , followed by GPT-5.6 Sol (77.4%) and GPT-5.6 Terra (76.7%). BenchLM does not use these results to rank models overall.

1Closed

Claude Opus 5

Anthropic

claude-opus-5

78.0%

Overall 82.81

2Closed

GPT-5.6 Sol

OpenAI

gpt-5-6-sol

77.4%

Overall 81.39Context 1M

3Closed

GPT-5.6 Terra

OpenAI

gpt-5-6-terra

76.7%

Overall 71.99Context 1M

92 modelsCodingCurrentDisplay onlyUpdated July 29, 2026

Benchmark score table (92 models)

Score

Claude Opus 5Anthropic · Closed

78.0%

GPT-5.6 SolOpenAI · Closed

77.4%

GPT-5.6 TerraOpenAI · Closed

76.7%

Claude Fable 5Anthropic · Closed

76.5%

Kimi K3Moonshot AI · Closed

76.2%

GPT-5.5OpenAI · Closed

74.9%

Claude Opus 4.8Anthropic · Closed

74.3%

Claude Opus 4.7 (Adaptive)Anthropic · Closed

73.6%

Grok 4.5xAI · Closed

72.5%

Claude Sonnet 5Anthropic · Closed

71.5%

GPT-5.6 LunaOpenAI · Closed

71.5%

Muse Spark 1.1Meta · Closed

71.3%

GPT-5.4OpenAI · Closed

71.0%

Gemini 3.5 FlashGoogle · Closed

70.1%

Gemini 3.6 FlashGoogle · Closed

69.2%

Gemini 3.1 ProGoogle · Closed

68.8%

GLM-5.2Z.AI · Open weight

68.8%

Qwen3.7 MaxAlibaba · Closed

66.0%

Kimi K2.6Moonshot AI · Open weight

61.8%

Kimi K2.7 CodeMoonshot AI · Open weight

60.8%

MiMo-V2.5-ProXiaomi · Closed

60.2%

DeepSeek V4 Pro (Max)DeepSeek · Open weight

59.4%

Hy3 PreviewTencent · Open weight

58.8%

Hy3Tencent · Open weight

58.8%

DeepSeek V4 Pro (High)DeepSeek · Open weight

58.7%

Muse SparkMeta · Closed

58.6%

MiniMax M3MiniMax · Open weight

58.6%

DeepSeek V4 Flash (Max)DeepSeek · Open weight

56.2%

GPT-5.4 miniOpenAI · Closed

56.1%

GPT-5.4 nanoOpenAI · Closed

56.1%

Qwen3.7 PlusAlibaba · Closed

55.9%

GLM-5.1Z.AI · Open weight

55.8%

Qwen3.6 PlusAlibaba · Closed

54.5%

Qwen3.6-27BAlibaba · Open weight

53.7%

MiniMax M2.7MiniMax · Open weight

52.6%

InklingThinking Machines Lab · Open weight

52.1%

DeepSeek V4 Flash (High)DeepSeek · Open weight

52.0%

MiMo-V2-FlashXiaomi · Open weight

49.8%

GPT-5.1OpenAI · Closed

49.4%

Gemini 3.5 Flash-LiteGoogle · Closed

49.3%

Nemotron 3 UltraNVIDIA · Open weight

49.3%

Qwen3.5 397BAlibaba · Open weight

48.2%

Qwen3.5 397B (Reasoning)Alibaba · Open weight

48.2%

Mistral Medium 3.5 128BMistral · Open weight

46.9%

Kimi K2.5Moonshot AI · Open weight

46.8%

Kimi K2.5 (Reasoning)Moonshot AI · Closed

46.8%

Qwen3.5-122B-A10BAlibaba · Open weight

45.7%

GLM-4.7Z.AI · Open weight

45.3%

Gemma 4 31BGoogle · Open weight

43.4%

Grok 4.3xAI · Closed

42.3%

Qwen3.6-35B-A3BAlibaba · Open weight

41.9%

o1OpenAI · Closed

39.7%

Step 3.7 FlashStepFun · Open weight

39.6%

Gemma 4 26B A4BGoogle · Open weight

39.3%

GPT-5 (high)OpenAI · Closed

37.8%

Nemotron 3 Super 120B A12BNVIDIA · Open weight

37.7%

Gemini 3.1 Flash-LiteGoogle · Closed

34.7%

o1-previewOpenAI · Closed

34.0%

Gemini 2.5 ProGoogle · Closed

33.3%

K-ExaoneLG AI Research · Closed

32.1%

Mercury 2Inception · Closed

31.1%

Gemma 4 12BGoogle · Open weight

31.0%

GPT-OSS 120BOpenAI · Open weight

30.4%

Command A+Cohere · Open weight

27.9%

Mistral Small 4Mistral · Open weight

26.6%

Mistral Small 4 (Reasoning)Mistral · Open weight

26.6%

Trinity-Large-PreviewArcee AI · Open weight

25.8%

Trinity-Large-ThinkingArcee AI · Open weight

25.8%

Ling 2.6 FlashInclusionAI · Open weight

25.3%

Gemini 1.5 ProGoogle · Closed

23.6%

DeepSeek V3DeepSeek · Open weight

23.0%

GPT-4 TurboOpenAI · Closed

21.5%

GPT-OSS 20BOpenAI · Open weight

20.7%

GPT-4.1 miniOpenAI · Closed

20.2%

Mistral Large 3Mistral · Closed

20.1%

Claude 3 OpusAnthropic · Closed

19.5%

Llama 4 MaverickMeta · Open weight

16.3%

GPT-5 miniOpenAI · Closed

15.6%

Nemotron 3 Nano 30BNVIDIA · Open weight

14.4%

Ministral 3 14B (Reasoning)Mistral · Open weight

14.4%

Ministral 3 14BMistral · Open weight

14.4%

Nemotron 3 Nano Omni 30B A3BNVIDIA · Open weight

13.8%

GPT-4o miniOpenAI · Closed

11.4%

GPT-4.1 nanoOpenAI · Closed

11.1%

Gemma 3 27BGoogle · Open weight

10.1%

Ministral 3 8B (Reasoning)Mistral · Open weight

9.7%

Ministral 3 8BMistral · Open weight

9.7%

Gemma 4 E4BGoogle · Open weight

9.4%

Llama 4 ScoutMeta · Open weight

8.2%

Gemma 4 E2BGoogle · Open weight

7.2%

Ministral 3 3B (Reasoning)Mistral · Open weight

4.8%

Ministral 3 3BMistral · Open weight

4.8%

The published AA Coding Index snapshot places Claude Opus 5 first at 78.0%. The third row is 1.3 points behind. The broader top-10 range is 6.4 points, so many of the published results sit in a relatively narrow band.

92 models have been evaluated on AA Coding Index. The benchmark falls in the Coding category. This category carries a 20% weight in BenchLM.ai's overall scoring system. AA Coding Index is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About AA Coding Index

Year

2026

Tasks

Cross-benchmark coding index

Format

Aggregated model score

Difficulty

Display-only external reference

BenchLM mirrors this coding index for comparison, but does not use it as a weighted coding benchmark row.

Artificial Analysis model leaderboards

BenchLM freshness & provenance

Version

AA Coding Index 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

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.

FAQ

What does AA Coding Index measure?

A display-only Artificial Analysis coding index.

Which model scores highest on AA Coding Index?

Claude Opus 5 by Anthropic currently leads with a score of 78.0% on AA Coding Index.

How many models are evaluated on AA Coding Index?

92 AI models have been evaluated on AA Coding Index on BenchLM.

Compare Top Models on AA Coding Index

Claude Opus 5 vs GPT-5.6 Sol GPT-5.6 Sol vs GPT-5.6 Terra GPT-5.6 Terra vs Claude Fable 5 Claude Fable 5 vs Kimi K3

Last updated: July 29, 2026 · BenchLM version AA Coding Index 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 issue

Join 2,000+ readers.

One email each week. Unsubscribe anytime.