AI ranking
Code
Solving real bugs (SWE-bench), file editing (Aider), long-form reasoning, reliability, cost.
Visual top 10
- 01 Deepseek r1 + claude 3 5 Sonnet 20241022 96.83
- 02 O3 (high) + gpt 4.1 95.80
- 03 Claude Opus 5 5 Max 92.19
- 04 GPT 6.1 sol (max) 89.48
- 05 Claude sonnet 5.5 (xhigh) 88.28
- 06 O3 (high) 85.83
- 07 GPT 5 (low) 85.70
- 08 Gemini 2.5 Pro Preview 06 05 (32k think) 85.04
- 09 o3 84.70
- 10 Claude Fable 5 1 Max 84.61
Compare models. Tick up to 3 models in the table below, then run the comparison: who wins on which use, by how much, and how solid the measurement is.
| # | Model | Vendor | Score | Monthly subscription | Assessment | |
|---|---|---|---|---|---|---|
| 01 | Anthropic | 96.83 / 100 | Free | Partial | ||
| 02 | OpenAI | 95.80 / 100 | €23/month | Partial | ||
| 03 | Anthropic | 92.19 / 100 | €23/month | Partial | ||
| 04 | OpenAI | 89.48 / 100 | €23/month | Reliable | ||
| 05 | Anthropic | 88.28 / 100 | Not available | Partial | ||
| 06 | OpenAI | 85.83 / 100 | €23/month | Partial | ||
| 07 | OpenAI | 85.70 / 100 | €23/month | Reliable | ||
| 08 | 85.04 / 100 | €22/month | Partial | |||
| 09 | OpenAI | 84.70 / 100 | €23/month | Reliable | ||
| 10 | Anthropic | 84.61 / 100 | €23/month | Partial | ||
| 11 | DeepSeek | 84.39 / 100 | Free | Partial | ||
| 12 | Anthropic | 84.33 / 100 | Not available | Partial | ||
| 13 | OpenAI | 84.23 / 100 | €23/month | Reliable | ||
| 14 | Meta | 83.78 / 100 | Not available | Partial | ||
| 15 | Anthropic | 83.65 / 100 | Not available | Partial | ||
| 16 | 83.27 / 100 | €22/month | Partial | |||
| 17 | OpenAI | 82.98 / 100 | €23/month | Partial | ||
| 18 | 82.81 / 100 | €22/month | Reliable | |||
| 19 | OpenAI | 82.48 / 100 | €23/month | Reliable | ||
| 20 | Anthropic | 82.45 / 100 | Not available | Partial | ||
| 21 | OpenAI | 82.30 / 100 | €23/month | Partial | ||
| 22 | 82.22 / 100 | €22/month | Partial | |||
| 23 | OpenAI | 82.15 / 100 | €23/month | Partial | ||
| 24 | OpenAI | 82.11 / 100 | €23/month | Partial | ||
| 25 | OpenAI | 82.10 / 100 | €23/month | Reliable | ||
| 26 | Anthropic | 82.09 / 100 | Not available | Partial | ||
| 27 | Anthropic | 81.82 / 100 | Not available | Partial | ||
| 28 | DeepSeek | 81.79 / 100 | Free | Partial | ||
| 29 | Spacexai | 81.64 / 100 | Free | Partial | ||
| 30 | xAI | 81.49 / 100 | Free | Partial | ||
| 31 | OpenAI | 81.48 / 100 | €23/month | Partial | ||
| 32 | OpenAI | 81.08 / 100 | €23/month | Reliable | ||
| 33 | OpenAI | 80.95 / 100 | €23/month | Partial | ||
| 34 | Anthropic | 80.94 / 100 | Not available | Partial | ||
| 35 | Anthropic | 80.69 / 100 | Not available | Reliable | ||
| 36 | OpenAI | 80.65 / 100 | €23/month | Partial | ||
| 37 | Spacexai | 80.51 / 100 | Free | Partial | ||
| 38 | Anthropic | 80.50 / 100 | Not available | Partial | ||
| 39 | 80.44 / 100 | €22/month | Partial | |||
| 40 | OpenAI | 80.05 / 100 | €23/month | Partial | ||
| 41 | Meta | 80.00 / 100 | Not available | Reliable | ||
| 42 | 79.99 / 100 | €22/month | Partial | |||
| 43 | OpenAI | 79.90 / 100 | €23/month | Partial | ||
| 44 | Anthropic | 79.79 / 100 | Not available | Partial | ||
| 45 | Anthropic | 79.78 / 100 | Not available | Partial | ||
| 46 | Moonshot AI | 79.77 / 100 | Not available | Partial | ||
| 47 | Anthropic | 79.53 / 100 | Not available | Partial | ||
| 48 | Alibaba | 79.27 / 100 | Not available | Reliable | ||
| 49 | Anthropic | 79.26 / 100 | Not available | Partial | ||
| 50 | Spacexai | 79.01 / 100 | Free | Partial |
Indicative score, weighted from public sources. Transparent methodology. See the AI ranking page for the full methodology and the other categories.
AI coding model ranking
Le Recul's Code ranking assesses AI models on what actually matters for development: fixing real bugs, editing existing files, holding a long line of reasoning across a codebase and staying reliable from one run to the next. It draws on public signals such as SWE-bench (solving real tickets) and Aider (code editing), weighted according to a « Code » grid.
Here too, there is no « best tool » in the abstract. A model can shine on isolated exercises and fail on a real codebase, or perform well but cost too much for intensive use. The ranking separates code models (Claude, GPT, Gemini, DeepSeek, Qwen Coder…) from agentic development systems (Claude Code, Cursor, Aider), which are handled separately under Coding agents.
The score reflects a trend measured on real tasks, not a commercial promise. Every evaluation carries a reliability status depending on source coverage, and the weightings stay public. Use it to choose a model for your context — language, project size, budget — rather than to follow the latest announcement.
What this ranking is for
This ranking helps you choose a model to assist development: completion, refactoring, bug fixing, test generation. It deliberately separates raw models from coding agents, which add a layer of tools and autonomy.
It also helps temper spectacular demos: succeeding on a scripted exercise does not guarantee reliability on a real repository, with its dependencies and its history.
How to read the ranking
- Solving real bugs (SWE-bench style), not toy exercises.
- Ability to edit existing files (Aider style).
- Long-form reasoning across a codebase, without losing the thread.
- Reliability and reproducibility from one run to the next.
- Cost relative to how intensively you use it.
FAQ — Code
What is the difference between a code model and a coding agent?
A code model generates and edits code as text. A coding agent (Claude Code, Cursor, Aider) adds tools, runs, tests and fixes in a loop. Agents are ranked separately under Coding agents.
What is the Code score based on?
Mainly on solving real bugs (SWE-bench) and code editing (Aider), completed by reasoning, reliability and cost, according to a documented « Code » grid.
Will a highly ranked model be reliable on my project?
The ranking gives a general trend. Your results depend on the language, the size of the repository and the context. Test on a representative task before concluding.
Are open-weights models included?
Yes, where the sources allow it. Open models that run locally also have a dedicated view (Local AI — code).
How often is this updated?
About every 25 hours, automatically, from public sources. The date is shown at the top of the page.