Best AI model for coding
Measured in Nivana’s weekly evals (updated 2026-10-25): Claude Opus scores highest on coding at 97/100. The router also weighs price and speed, so its pick depends on how hard the request is.
What the router picks
- Easy: GPT-6 Luna
- Medium: Claude Sonnet
- Hard: Claude Sonnet
Ranked by measured quality
| Model | Lab | Score | Price |
|---|---|---|---|
| Claude Opus | Anthropic | 97 | $4 in / $20 out per 1M tokens |
| Claude Sonnet | Anthropic | 94 | $2 in / $10 out per 1M tokens |
| GPT-5.5 | OpenAI | 90 | $5 in / $30 out per 1M tokens |
| GPT-6.1 Sol | OpenAI | 88 | $2 in / $10 out per 1M tokens |
| Kimi K3.1 Pro | Moonshot | 87 | $2 in / $10 out per 1M tokens |
| DeepSeek V4 Pro | DeepSeek | 85 | $0.66 in / $1.98 out per 1M tokens |
| Gemini 4 Argon | 85 | $2 in / $10 out per 1M tokens | |
| DeepSeek Flash | DeepSeek | 85 | $0.075 in / $0.3 out per 1M tokens |
| Muse Spark | Meta | 84 | $1.25 in / $4.25 out per 1M tokens |
| Grok 4.7 | xAI | 83 | $2 in / $6 out per 1M tokens |
| GPT-6 Luna | OpenAI | 81 | $0.1 in / $0.5 out per 1M tokens |
| Gemini Flash | 79 | $0.75 in / $3.75 out per 1M tokens | |
| Mistral Large | Mistral AI | 79 | $1.36 in / $4.18 out per 1M tokens |
| Claude Haiku | Anthropic | 78 | $0.1 in / $0.5 out per 1M tokens |