Best open-source models for coding
The best open-source or open-weight coding models are ranked here by SWE-bench score per effective dollar, then compared by price, context, and host availability. The current classification is provisional: verify each model's licence before commercial use.
| # | Model | SWE-bench | Cheapest host | Input $/MTok | Output $/MTok | Hosts | Context |
|---|---|---|---|---|---|---|---|
| 1 | Kimi K2.5 | 73.8% | TensorX | $0.50 | $2.80 | 7 | — |
| 2 | Kimi K2.6 | 76.7% | Inceptron | $0.73 | $3.50 | 3 | 262k |
| 3 | DeepSeek V4 Pro 0423 | 77.6% | Fireworks AI | $1.32 | $3.48 | 5 | 1.048576M |
| 4 | GLM 5.1 | 74.2% | Fireworks AI | $1.40 | $4.40 | 1 | — |
How this ranking works
Models must have a benchmark score at or above the site's coding threshold. The ranking uses SWE-bench divided by effective blended token cost. Host prices are compared separately; a cheaper host does not imply that the model itself is licensed for commercial use.
Open-source and open-weight are not interchangeable legal categories. Check the model licence and the provider's hosting terms before using any model in a commercial product.