Category ranking · Coding & agents

Best LLMs for Coding

Software engineering is where model differences are most visible. This ranking weights agentic coding, tool use, SWE-bench and terminal-task performance — the dimensions that matter for real codebases.

Benchmarks rotate and vendors use different harnesses. Treat this as a shortlist, then run your own evaluation. Data as of 2026-11-20.
# Model CodingOverallReasoningValuePrice (in/out)License
1 Claude Opus 5.5
Anthropic · US
9993.39770$4 / $20Proprietary
2 Kimi K3.1 Pro
Moonshot AI · CN
9389.89082$2 / $10Open weights
3 Claude Sonnet 5.5
Anthropic · US
9288.98790$2 / $10Proprietary
4 GPT-6 Astra
OpenAI · US
9190.19662$10 / $50Proprietary
5 GLM-5.3
Z.ai (Zhipu) · CN
8984.58390$0.60 / $2.20Open weights
6 GPT-6.1 Sol
OpenAI · US
8889.38992$2 / $10Proprietary
7 DeepSeek V4.1 Flash
DeepSeek · CN
8883.68099$0.15 / $0.60Open source
8 Grok 4.7
xAI · US
8784.38486$2 / $6Proprietary
9 Gemini 4 Argon
Google · US
8689.78884$2 / $10Proprietary
10 Qwen3.8 Max
Alibaba Cloud · CN
8688.88788$2 / $6Proprietary
11 Mistral Large 4
Mistral AI · EU
8080.57988$0.70 / $2.10Open weights
12 Doubao Seed 2.1 Pro
ByteDance · CN
7882.78094$0.30 / $1.20Proprietary
13 Llama 4 Maverick
Meta · US
7678.57590$0.20 / $0.60Open source