Meta-Llama
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| LLM Inference | Meta-Llama-3-8B prompt/gen = 128 tokens | Memory Usage (MB)5,463 | 12 | |
| Model provenance detection | Meta-Llama-3-8B Quantization | TPR83.3 | 5 | |
| Model provenance detection | Meta-Llama-3-8B Post-training | TPR94.4 | 5 | |
| Language Model Inference Efficiency | Meta-Llama-3-8B | Throughput (tokens/s)1,398 | 4 |
Showing 4 of 4 rows