LEADERBOARD
Coding model leaderboard
View model performance, conservative rating, open-weight status, and minimum input price from the LLM Stats coding category. The ranking is reference data, not a single product recommendation.
Data sourceLLM Stats
Ranking methodtrueskill
Ranked atSep 29, 2026, 1:36 PM
Models included50
How to read the overall board
The overall board aggregates several coding benchmarks. Use it to compare broad relative position, then inspect individual tasks, price, and deployment constraints.
| Rank | Model | Organization | Score | Conservative | License | Min input price | Evaluated |
|---|---|---|---|---|---|---|---|
| 1 | anthropic | 0.7 | 51.6 | Closed | $4 | 12 | |
| 2 | anthropic | 0.7 | 48.9 | Closed | $2 | 12 | |
| 3 | openai | 0.7 | 48.7 | Closed | $10 | 8 | |
| 4 | openai | 0.6 | 45.1 | Closed | $5 | 17 | |
| 5 | anthropic | 0.7 | 44.6 | Closed | $10 | 9 | |
| 6 | anthropic | 0.8 | 43.9 | Closed | — | 6 | |
| 7 | deepseek | 0.5 | 43.1 | Open source | $0.22 | 9 | |
| 8 | moonshotai | 0.7 | 42.0 | Open weight | $2.85 | 9 | |
| 9 | zai-org | 0.5 | 41.5 | Open weight | $1.2 | 12 | |
| 10 | openai | 0.6 | 41.4 | Closed | $2 | 17 | |
| 11 | meta | 0.7 | 41.2 | Closed | $0.1 | 3 | |
| 12 | xiaomi | 0.6 | 41.1 | Open source | $0.435 | 9 | |
| 13 | anthropic | 0.6 | 40.9 | Closed | $5 | 4 | |
| 14 | anthropic | 0.7 | 40.1 | Closed | $5 | 11 | |
| 15 | deepseek | 0.7 | 39.3 | Open source | $1.3 | 5 |