Agents that read a codebase, edit files and run tests. Long prompts, long answers and many steps per task.
8 models meet the requirements, ranked by monthly cost for this workload.
| Model | Arena score | Class | Context | Images | Input $/1M | Output $/1M | Monthly |
|---|---|---|---|---|---|---|---|
| DeepSeek V4 Pro Lowest cost | 1,465 | Reasoning | 1M | — | $0.66 | $1.98 | $176 |
| Grok 4.7 | 1,441 | Flagship | 500K | ✓ | $2.00 | $6.00 | $612 |
| Claude Sonnet 5.5 | 1,477 | Balanced | 1M | ✓ | $2.00 | $10.00 | $648 |
| GPT-6.1 Sol | 1,481 | Balanced | 1.05M | ✓ | $2.00 | $10.00 | $648 |
| Gemini 4 Argon Promo | 1,520 | Flagship | 1.05M | ✓ | $2.00 | $10.00 | $648 |
| Claude Opus 5.5 | 1,509 | Flagship | 1M | ✓ | $4.00 | $20.00 | $1,296 |
| Claude Fable 5.1 | 1,504 | Frontier | 1M | ✓ | $10.00 | $50.00 | $3,195 |
| GPT-6 Astra | 1,476 | Frontier | 1.05M | ✓ | $10.00 | $50.00 | $3,330 |
Arena score: LMArena (text, style control), from human votes, published Oct 28, 2026, CC BY 4.0.
Our side-by-side tests for this task are in progress. Until they are published, run 10 to 20 of your own real examples on two or three candidates before you choose.
Independent leaderboards: LMArena · Artificial Analysis
How this list is built: models in the classes suited to the task that meet its hard requirements (image input, context size, open weights), ranked by price. A lower price does not mean better quality.