Grammatical Understanding on MIXED (test)
95Task 1 AccuracyQwen/Qwen3-Next-80B-A3B-Thinking
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Qwen/Qwen3-Next-80B-A3B-ThinkingParameters=80B, Setting=Base Settings2025.10 | 95 | 95 | 91 | 58 | |
| GPT-5Setting=Base Settings2025.10 | 94 | 95 | 91 | 61 | |
| Qwen/Qwen3-Next-80B-A3B-InstructParameters=80B, Setting=Base Settings2025.10 | 92 | 95 | 88 | 57 | |
| google/gemma-3-27b-itParameters=27B, Setting=Base Settings2025.10 | 88 | 95 | 87 | 54 | |
| meta-llama/Llama-3.2-1B-InstructParameters=1B, Setting=Base Settings2025.10 | 88 | 59 | 43 | 51 | |
| meta-llama/Llama-3.3-70B-InstructParameters=70B, Setting=Base Settings2025.10 | 88 | 95 | 84 | 51 | |
| google/gemma-3-12b-itParameters=12B, Setting=Base Settings2025.10 | 87 | 95 | 85 | 54 | |
| Qwen/Qwen3-4B-Instruct-2507Parameters=4B, Setting=Base Settings2025.10 | 86 | 90 | 82 | 52 | |
| Qwen/Qwen3-30B-A3B-Instruct-2507Parameters=30B, Setting=Base Settings2025.10 | 85 | 95 | 81 | 54 | |
| meta-llama/Llama-3.2-3B-InstructParameters=3B, Setting=Base Settings2025.10 | 84 | 75 | 62 | 47 | |
| google/gemma-2-9b-itParameters=9B, Setting=Base Settings2025.10 | 83 | 92 | 83 | 51 | |
| google/gemma-3-4b-itParameters=4B, Setting=Base Settings2025.10 | 81 | 79 | 47 | 52 | |
| google/gemma-2-2b-itParameters=2B, Setting=Base Settings2025.10 | 80 | 87 | 60 | 50 |