Every model we’ve tested, & your votes
This is to help you compare. Remember there is no perfect model, just different model biases for different tasks.
| Provider | Model | AVG Score | Times Tested | Properties |
|---|---|---|---|---|
|
Claude 3 7 Sonnet 20250219 V1:0 | 40% | 2 |
|
|
|
Claude 3 Sonnet 20240229 V1:0 | 0% | 1 |
|
|
|
Claude Opus 4 6 V1 | 0% | 2 |
|
|
|
Claude Sonnet 4 6 | 14% | 2 |
|
|
Gpt Oss 120B 1:0 | 0% | 2 |
|
|
|
Gpt Oss 20B 1:0 | 0% | 2 |
|
| Moonshotai | Kimi K2.5 | 0% | 2 |
|
| Minimax | Minimax M2.1 | 0% | 2 |
|
|
Mistral Large 2402 V1:0 | 7% | 2 |
|
| Amazon | Nova Lite V1:0 | 0% | 2 |
|
| Amazon | Nova Pro V1:0 | 0% | 2 |
|
| Qwen | Qwen3 235B A22B 2507 V1:0 | 0% | 2 |
|
| Qwen | Qwen3 32B V1:0 | 0% | 2 |
|
| Qwen | Qwen3 Coder 30B A3B V1:0 | 17% | 2 |
|
| Qwen | Qwen3 Coder 480B A35B V1:0 | 0% | 1 |
|
| Qwen | Qwen3 Coder Next | 0% | 1 |
|
|
V3.2 | 14% | 1 |
|
|
|
V3 V1:0 | 29% | 1 |
|