DeepSeek-R1-Distill-Llama
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Multi-step LLM Interaction | DeepSeek-R1-Distill-Llama-8B Inference Performance (test) | High-Precision Ratio100 | 7 | |
| End-to-end single-step decoding | DeepSeek-R1-Distill-LLaMA-8B 32K Context | Latency (ms)23.4 | 4 |
Showing 2 of 2 rows