Leaderboard
Aider Polyglot
26 models scored · metric: percent correct. Prices are the cheapest listed across all serving providers. Pts/$ = score ÷ blended price (3×input + output) / 4 — how much score each dollar buys.
Best value · top 5 by pts per dollar
- 1.DeepSeek Reasoner424
- 2.DeepSeek Chat401
- 3.Gemini 2.5 Flash90.7
- 4.GPT-4.1 nano56.5
- 5.GPT-4.1 mini52.3
| # | Model | Score | Best in /M | Best out /M | Pts / $ | Context |
|---|---|---|---|---|---|---|
| 🥇 | GPT-5openai | 88 | $1.10 | $9 | 28.6 | 400K |
| 🥈 | o3-proopenai | 84.9 | $18 | $72 | 2.7 | 200K |
| 🥉 | Gemini 2.5 Progoogle | 83.1 | $0.87 | $7 | 34.6 | 1.05M |
| 4 | o3openai | 81.3 | $1.80 | $7.20 | 25.8 | 200K |
| 5 | DeepSeek Reasonerdeepseek | 74.2 | $0.14 | $0.28 | 424 | 1M |
| 6 | Claude Opus 4anthropic | 72 | $15 | $75 | 2.4 | 200K |
| 7 | o4-miniopenai | 72 | $0.99 | $4 | 41.3 | 200K |
| 8 | DeepSeek Chatdeepseek | 70.2 | $0.14 | $0.28 | 401 | 1M |
| 9 | Claude Sonnet 3.7anthropic | 64.9 | $3 | $15 | 10.8 | 200K |
| 10 | o1openai | 61.7 | $14 | $54 | 2.6 | 200K |
| 11 | Claude Sonnet 4anthropic | 61.3 | $3 | $15 | 10.2 | 200K |
| 12 | o3-miniopenai | 60.4 | $0.99 | $4 | 34.7 | 200K |
| 13 | Qwen3 235B-A22Balibaba | 59.6 | $0.7 | $2.80 | 48.7 | 131K |
| 14 | DeepSeek-R1deepseek | 56.9 | $0.7 | $2.50 | 49.5 | 128K |
| 15 | Gemini 2.5 Flashgoogle | 55.1 | $0.21 | $1.80 | 90.7 | 1.05M |
| 16 | GPT-4.1openai | 52.4 | $1.80 | $7.20 | 16.6 | 1.05M |
| 17 | Qwen3 32Balibaba | 40 | $0.7 | $2.80 | 32.7 | 131K |
| 18 | GPT-4.1 miniopenai | 32.4 | $0.36 | $1.40 | 52.3 | 1.05M |
| 19 | GPT-4oopenai | 23.1 | $1.25 | $5 | 10.6 | 128K |
| 20 | GPT-4o (2024-08-06)openai | 23.1 | $2.50 | $10 | 5.3 | 128K |
| 21 | Qwen Maxalibaba | 21.8 | $1.60 | $6.40 | 7.8 | 33K |
| 22 | GPT-4o (2024-11-20)openai | 18.2 | $2.50 | $10 | 4.2 | 128K |
| 23 | Command Acohere | 12 | $2.50 | $10 | 2.7 | 256K |
| 24 | Codestral (latest)mistral | 11.1 | $0.3 | $0.9 | 24.7 | 256K |
| 25 | GPT-4.1 nanoopenai | 8.9 | $0.09 | $0.36 | 56.5 | 1.05M |
| 26 | GPT-4o miniopenai | 3.6 | $0.075 | $0.3 | 27.4 | 128K |
Cheapest scorer: GPT-4o mini at $0.075 input /M.