Compare coding models
Price, context, and public bench ranks — two or three models, no composite score.
| Attribute | Gemini 3.7 FlashGoogle | GLM-5.3Zhipu AI | Claude Sonnet 5Anthropic |
|---|---|---|---|
| Input | $0.75 | $1.40 | $3.00 |
| Output | $3.75 | $4.40 | $15.00 |
| Context | 1.0M | 1M | 1M |
| Status | error | error | verified |
| Tags | new, flagship, coding, fast, hybrid-thinking, reasoning, multimodal | new, flagship, coding, reasoning, long-context, agent, always-thinking | balanced, coding, vision, adaptive-thinking |
| CursorBench | #3 | #16 | #13 |
| LMArena Agent | #— | #— | #— |
| Artificial Analysis | #10 | #32 | #12 |
| Aider Polyglot | #— | #— | #— |