Scientific Question Answering on ARC-C (Runtime Performance)
1,163Latency (NoKV, ms)NoKV
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| NoKVModel=Qwen2.5-7B Instruct2026.06 | 1,163 | — | — | — | — | — | |
| NoKVModel=Falcon3-7B Instruct2026.06 | 1,173 | — | — | — | — | — | |
| NoKVModel=Mistral-7B Instruct2026.06 | 1,263 | — | — | — | — | — | |
| NoKVModel=Llama-3-8B Instruct2026.06 | 1,282 | — | — | — | — | — | |
| NoKVModel=Falcon3-10B Instruct2026.06 | 1,628 | — | — | — | — | — | |
| RKSCModel=Qwen2.5-7B Instruct2026.06 | — | — | 391 | — | 2.97 | 71 | |
| RKSCModel=Mistral-7B Instruct2026.06 | — | — | 482 | — | 2.62 | 60 | |
| RKSCModel=Falcon3-7B Instruct2026.06 | — | — | 347 | — | 3.38 | 66 | |
| RKSCModel=Falcon3-10B Instruct2026.06 | — | — | 418 | — | 3.89 | 77 | |
| RKSCModel=Llama-3-8B Instruct2026.06 | — | — | 401 | — | 3.2 | 74 | |
| vLLMModel=Qwen2.5-7B Instruct2026.06 | — | 651 | — | 1.79 | — | — | |
| vLLMModel=Mistral-7B Instruct2026.06 | — | 691 | — | 1.83 | — | — | |
| vLLMModel=Falcon3-7B Instruct2026.06 | — | 651 | — | 1.8 | — | — | |
| vLLMModel=Falcon3-10B Instruct2026.06 | — | 897 | — | 1.82 | — | — | |
| vLLMModel=Llama-3-8B Instruct2026.06 | — | 712 | — | 1.8 | — | — |