Qwen/Qwen3-4B-AWQ
| benchmark | category | score | context | conf | evidence |
|---|---|---|---|---|---|
| MMLU | knowledge | 83.7 | low | | Thinking | bf16 | 63.6 | 55.9 | 83.7 | 73.8 | | |
| AIME 2024 | math | 73.8 | low | | Thinking | bf16 | 63.6 | 55.9 | 83.7 | 73.8 | | |
| GPQA | reasoning | 55.9 | low | | Thinking | bf16 | 63.6 | 55.9 | 83.7 | 73.8 | |
none scraped.