Best model for your job
Best AI model for coding
For coding, Claude Fable 5.1 scores highest on the current data (69.4), weighting coding 60%, agentic & tool use 25%, instruction following 15%. The cheapest model in the top 10 is Claude Sonnet 5.5 at $2 / $10 per million tokens.
Last verified
Weights: Coding 60%, Agentic & Tool Use 25%, Instruction Following 15%. Coding work is mostly repository-level bug fixing and multi-step edits, so coding and agentic results carry the most weight.
| # | Model | Provider | Fit | Coding | Agentic & Tool Use | Instruction Following | Input $/M | Output $/M |
|---|---|---|---|---|---|---|---|---|
| 1 | Claude Fable 5.1 | Anthropic | 69.4 | 74.7 | 50.7 | 79.2 | $10 | $50 |
| 2 | GPT-6 Astra | OpenAI | 68.9 | 73.7 | 52.9 | 76.3 | $10 | $50 |
| 3 | Claude Fable 5 | Anthropic | 67.7 | 70.6 | 54.0 | 78.6 | $10 | $50 |
| 4 | Claude Opus 5.5 | Anthropic | 66.5 | 71.9 | 45.3 | 80.0 | $4 | $20 |
| 5 | Claude Opus 5 | Anthropic | 66.3 | 67.5 | 55.6 | 79.2 | $5 | $25 |
| 6 | Claude Sonnet 5.5 | Anthropic | 63.3 | 67.3 | 45.0 | 78.3 | $2 | $10 |
| 7 | GPT-5.6 Sol | OpenAI | 63.3 | 65.1 | 50.3 | 77.7 | $4 | $20 |
| 8 | Gemini 4 Argon | 61.5 | 62.0 | 49.2 | 80.1 | — | — | |
| 9 | Claude Opus 4.7 | Anthropic | 59.5 | 59.6 | 47.9 | 78.4 | $5 | $25 |
| 10 | Claude Opus 4.8 | Anthropic | 59.5 | 59.9 | 47.6 | 77.4 | $5 | $25 |
| 11 | GPT-6.1 Sol | OpenAI | 59.4 | 63.2 | 39.6 | 77.0 | $2 | $10 |
| 12 | GPT-5.5 | OpenAI | 59.2 | 58.2 | 50.7 | 77.5 | $5 | $30 |
| 13 | Claude Opus 4.6 | Anthropic | 59.0 | 57.2 | 51.1 | 79.5 | $5 | $25 |
| 14 | Kimi K3 (open weights) | Moonshot AI | 58.7 | 61.0 | 41.8 | 77.7 | $3 | $15 |
| 15 | Gemini 3.8 Flash | 57.7 | 59.2 | 41.8 | 78.0 | $0.75 | $3.75 | |
| 16 | GPT-6 Sol | OpenAI | 56.6 | 60.1 | 37.2 | 74.5 | $2 | $10 |
| 17 | GLM-5.3 (open weights) | Z.ai (Zhipu) | 56.4 | 59.5 | 36.4 | 77.5 | $1.40 | $4.40 |
| 18 | Claude Opus 4.5 | Anthropic | 56.3 | 54.8 | 47.3 | 77.5 | $5 | $25 |
| 19 | Grok 4.6 | xAI | 56.2 | 58.5 | 39.4 | 75.4 | $2 | $6 |
| 20 | GPT-5.6 Terra | OpenAI | 56.1 | 57.7 | 40.1 | 76.4 | $2 | $12 |
| 21 | Gemini 3.7 Flash | 55.9 | 56.2 | 42.1 | 77.7 | $0.75 | $3.75 | |
| 22 | Claude Sonnet 5 | Anthropic | 55.4 | 55.5 | 42.8 | 76.3 | $2 | $10 |
| 23 | Muse Spark 1.3 | Meta | 55.3 | 56.6 | 38.6 | 77.5 | $1.25 | $4.25 |
| 24 | Qwen3.8 Max | Alibaba (Qwen) | 55.1 | 53.5 | 45.4 | 77.6 | $2 | $6 |
| 25 | Grok 4.7 | xAI | 55.1 | 58.0 | 36.7 | 74.1 | $2 | $6 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Top two head to head: Claude Fable 5.1 vs GPT-6 Astra
Frequently asked questions
What is the best ai model for coding?
For coding, Claude Fable 5.1 scores highest on the current data (69.4), weighting coding 60%, agentic & tool use 25%, instruction following 15%. The cheapest model in the top 10 is Claude Sonnet 5.5 at $2 / $10 per million tokens.
How is this shortlist built?
Coding work is mostly repository-level bug fixing and multi-step edits, so coding and agentic results carry the most weight. Each model's category scores are blended with those weights; only ranked models with results in every needed category are listed.
Other jobs
- Best AI model for debugging
- Best AI model for building websites
- Best local LLM for coding
- Best AI model for agents
- Best AI model for RAG
- Best AI model for research
- Best AI model for studying
- Best AI model for math
- Best AI model for hard reasoning
- Best AI model for data analysis
- Best AI model for Excel
- Best AI model for data extraction
- Best AI model for writing
- Best AI model for emails
- Best AI model for copywriting
- Best AI model for creative writing
- Best AI model for translation
- Best AI model for language learning
- Best AI model for customer service
- Best AI model for presentations
- Best AI model for PDFs
- Best AI assistant for everyday use