Instruction Following on CodaSet ID IFEVAL (test)
88.76AccuracyDeepSeek-V3.1
Evaluation results
| Method | Links | ||
|---|---|---|---|
| DeepSeek-V3.12026.05 | 88.76 | 4.9 | |
| gpt-oss-120b2026.05 | 87.56 | 3.1 | |
| DecoRfallback_model=DeepSeek-V3.12026.05 | 87.56 | 4.5 | |
| DecoRfallback_model=gpt-oss-120b2026.05 | 87.32 | 3 | |
| Qwen3-235B-A22B2026.05 | 86.84 | 3.3 | |
| EmbedLLM2026.05 | 86.84 | 3.2 | |
| DeepSeek-V3.2-Exp2026.05 | 86.6 | 2.4 | |
| RouterDC2026.05 | 85.41 | 5 | |
| MODEL-SAT2026.05 | 85.41 | 3.5 | |
| Kimi-K2-Instruct2026.05 | 85.17 | 12.8 | |
| gemma-3-27b-it2026.05 | 84.69 | 1.4 | |
| LLM Router2026.05 | 84.19 | 3.6 | |
| Random Router2026.05 | 82.93 | 3 | |
| KNN Router2026.05 | 82.78 | 3.3 | |
| gemma-3-12b-it2026.05 | 81.82 | 1.2 | |
| Mistral-Small-3.2-24B2026.05 | 77.03 | 1 |