Holistic Evaluation on CodaSet ID Average (test)
90.6AccuracyQwen3-235B-A22B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Qwen3-235B-A22B2026.05 | 90.6 | 5 | |
| DeepSeek-V3.12026.05 | 89.74 | 5.2 | |
| DecoRfallback_model=DeepSeek-V3.12026.05 | 89.35 | 2.1 | |
| DecoRfallback_model=gpt-oss-120b2026.05 | 89.13 | 1.9 | |
| gpt-oss-120b2026.05 | 88.71 | 2.7 | |
| DeepSeek-V3.2-Exp2026.05 | 88.51 | 3.9 | |
| Kimi-K2-Instruct2026.05 | 87.94 | 11.9 | |
| MODEL-SAT2026.05 | 87.5 | 7.8 | |
| EmbedLLM2026.05 | 86.8 | 4.2 | |
| RouterDC2026.05 | 86.77 | 4.2 | |
| KNN Router2026.05 | 86.6 | 2 | |
| Random Router2026.05 | 84.18 | 4.4 | |
| gemma-3-27b-it2026.05 | 83.79 | 2.4 | |
| gemma-3-12b-it2026.05 | 82.83 | 1 | |
| LLM Router2026.05 | 82.23 | 2.3 | |
| Mistral-Small-3.2-24B2026.05 | 72.39 | 1.8 |