Free models for coding
Dedicated coding models (Qwen Coder, Codestral, Devstral, Laguna) and strong general models that code well: GLM, Kimi, DeepSeek, MiniMax. They plug into Cursor, Cline, Kilo Code, OpenCode, and Aider.
57 models10 providersup to 1M contextbest: GLM 5.3 FlashUpdated:
Read more
Large repositories need context, autocomplete needs low latency, and agentic coding needs reliable tool calling.
| # | Model ↕ | AA Index ↕ | GPQA ↕ | Context ↕ | Size ↕ | Free at ↕ | |
|---|---|---|---|---|---|---|---|
| 1 |  |
41.8 | 91% | 1M | Mid-size |     OrcaRouter, NVIDIA Build, UnoRouter, OpenCode Zen |
|
| 2 | |
44.8 | 92% | 1M | Frontier |  Together AI |
|
| 3 |  |
39.5 | — | 1M | Frontier | OrcaRouter, NVIDIA Build, OpenCode Zen |
|
| 4 |  |
40.9 | 95% | 1M | Mid-size | Google AI Studio |
|
| 5 |  |
33.7 | 91% | 262K | Mid-size |   Groq, Cloudflare Workers AI, UnoRouter |
|
| 6 | |
34.3 | 91% | 1M | Frontier | OrcaRouter |
|
| 7 | |
34.0 | 93% | 1M | Mid-size | UnoRouter |
|
| 8 | |
33.7 | 90% | 1M | Frontier | Together AI |
|
| 9 |  |
27.0 | 91% | 256K | Frontier | NVIDIA Build |
|
| 10 |  |
22.9 | 87% | 1M | Frontier |  OpenRouter, NVIDIA Build, UnoRouter |
|
| 11 | |
22.2 | 84% | 1M | Small | UnoRouter |
|
| 12 |  |
22.8 | 85% | 205K | Frontier | UnoRouter |
|
| 13 | |
21.4 | 86% | 262K | Frontier | UnoRouter |
|
| 14 | |
21.4 | 84% | 262K | Mid-size | UnoRouter |
|
| 15 | |
22.2 | 86% | 203K | Frontier | Together AI |
|
| 16 | |
18.2 | 84% | 262K | Mid-size | Together AI, UnoRouter |
|
| 17 | |
19.3 | 85% | 262K | Mid-size | Together AI |
|
| 18 | |
17.7 | 83% | 262K | Frontier | Together AI |
|
| 19 | |
18.6 | 78% | 197K | Frontier | Together AI |
|
| 20 | |
— | — | 1M | Frontier | NVIDIA Build, UnoRouter |
|
| 21 | |
14.9 | 58% | 131K | Mid-size | Cloudflare Workers AI, UnoRouter |
|
| 22 | |
13.7 | 81% | 262K | Small | Together AI, UnoRouter |
|
| 23 | |
12.8 | 80% | 262K | Frontier | NVIDIA Build, UnoRouter |
|
| 24 |  |
11.6 | 78% | 131K | Frontier | Groq, Cloudflare Workers AI, UnoRouter |
|
| 25 | |
13.1 | 77% | 262K | Small | UnoRouter |
|
| 26 | |
— | — | 203K | Frontier | Together AI |
|
| 27 | |
— | — | 203K | Frontier | Together AI |
|
| 28 | |
— | — | 8K | Frontier | Together AI |
|
| 29 | |
12.8 | 80% | 262K | Frontier | Together AI |
|
| 30 | |
12.7 | 79% | 262K | Frontier | Together AI |
|
| 31 |  |
9.9 | 76% | 262K | Small | OpenRouter, UnoRouter |
|
| 32 | |
11.2 | 76% | 262K | Frontier | Together AI |
|
| 33 | |
— | — | 1M | Mid-size | UnoRouter |
|
| 34 | |
9.6 | 52% | 262K | Frontier | Together AI |
|
| 35 | |
— | — | 262K | Mid-size | UnoRouter |
|
| 36 | — | — | 131K | Mid-size | Together AI |
||
| 37 | |
— | — | 131K | Mid-size | UnoRouter |
|
| 38 | |
— | — | 128K | Mid-size | UnoRouter |
|
| 39 |  |
8.7 | 43% | 131K | Mid-size | Together AI |
|
| 40 | |
6.7 | 42% | 33K | Mid-size | Cloudflare Workers AI, UnoRouter |
|
| 41 | |
6.9 | 46% | 262K | Small | Together AI |
|
| 42 | |
7.6 | 68% | 66K | Frontier | Together AI |
|
| 43 | — | — | 262K | Small | NVIDIA Build, UnoRouter |
||
| 44 | — | — | 1M | Small | UnoRouter |
||
| 45 | |
— | — | 256K | Small | Mistral AI |
|
| 46 |  |
— | — | — | Frontier | NVIDIA Build |
|
| 47 | |
— | — | 256K | Small | UnoRouter |
|
| 48 |
SWE 1 6 Slow
Coding
|
— | — | 6K | Small | UnoRouter |
|
| 49 |
Voyage Code 2
Coding
|
— | — | 16K | Small | UnoRouter |
|
| 50 |
Voyage Code 4
Coding
|
— | — | 32K | Small | UnoRouter |
|
| 51 | |
— | — | — | Mid-size | NVIDIA Build |
|
| 52 | — | — | — | Mid-size | NVIDIA Build |
||
| 53 | — | — | — | Mid-size | NVIDIA Build |
||
| 54 |  |
— | — | — | Small | NVIDIA Build |
|
| 55 | |
— | — | — | Small | NVIDIA Build |
|
| 56 | |
— | — | — | Small | NVIDIA Build |
|
| 57 | — | — | — | Small | NVIDIA Build |
No models match your filters.
Benchmarks: Artificial Analysis (September 24, 2026). A dash means the model is not on their leaderboard; we never estimate scores ourselves.
FAQ
What is the best free model for coding?
GLM 5.3 Flash currently leads — free at OrcaRouter, NVIDIA Build, UnoRouter, OpenCode Zen. Next come GLM 5.3, DeepSeek V4.1 Flash, Gemini 3.8 Flash.