💻 Best AI Models for Coding
Ranking of language models by coding performance across multiple benchmarks including code generation, debugging, refactoring, and software engineering tasks.Updated August 2026. Based on independent benchmark data.
Why This Matters
Choosing the right model for coding is about more than raw benchmark scores. The models below are ranked using a weighted blend of category-specific benchmarks, arena Elo, and the intelligence index — but real-world performance depends on your context window, latency budget, and integration path.
For coding, we weighted the benchmarks that most strongly predict success in production: task accuracy first, then speed and cost efficiency. Leaders on this page consistently outperform on both standardized tests and real-world workloads.
If you operate under a strict budget, check the Best Value LLMs ranking. If latency matters more than raw quality, see Fastest LLMs. Both pages use the same underlying dataset with different weightings.
| # | Model | Provider | Score | Speed | Price (output) |
|---|---|---|---|---|---|
| 1 | Claude Fable 5 | Anthropic | 84.1 | 71 t/s | $50/M |
| 2 | Claude Mythos Preview | Anthropic | 80.9 | 80 t/s | $15/M |
| 3 | Claude Opus 5 | Anthropic | 70.4 | 60 t/s | $25/M |
| 4 | Claude Mythos 5 | Anthropic | 66.0 | 50 t/s | $50/M |
| 5 | GPT-5.6 Sol | OpenAI | 64.8 | 72 t/s | $30/M |
| 6 | Muse Spark 1.1 | Meta | 64.4 | 208 t/s | $4.25/M |
| 7 | Kimi K3 | Moonshot AI | 61.5 | 37 t/s | $15/M |
| 8 | Gemini 3.6 Flash | 60.2 | 233 t/s | $7.5/M | |
| 9 | Claude Opus 4.8 | Anthropic | 58.0 | 50 t/s | $15/M |
| 10 | GLM-5.1 | Z.ai | 57.0 | 74 t/s | $4.4/M |
| 11 | Claude Sonnet 4.6 (max) | Anthropic | 57.0 | 50 t/s | $15/M |
| 12 | GPT-5.3 Codex | OpenAI | 55.5 | 118 t/s | $14/M |
| 13 | MiniMax-M2.7 | MiniMax | 55.0 | 40 t/s | $2/M |
| 14 | GPT-5.6 Terra | OpenAI | 54.9 | 127 t/s | $12/M |
| 15 | GPT-5.5 | OpenAI | 54.0 | 60 t/s | $10/M |
| 16 | Qwen3.8 Max | Alibaba | 50.3 | 55 t/s | $2/M |
| 17 | Gemini 3.5 Flash | 49.6 | 267 t/s | $9/M | |
| 18 | Claude Opus 4.7 | Anthropic | 49.2 | 80 t/s | $15/M |
| 19 | GPT-5.6 Luna | OpenAI | 45.6 | 175 t/s | $1.2/M |
| 20 | Grok 4.5 | xAI | 45.1 | 59 t/s | $6/M |
Top 5 Coding — Quick Comparison
Methodology
Rankings are based on independent benchmark evaluations. The coding score is computed from multiple standardized benchmarks and normalized to 0–100. Data is updated regularly from public benchmark results.