Increase Llama 2's Latency and Throughput Performance by Up to 4X | Towards Data Science

Real-world benchmarks for Llama-2 13B

By · · 1 min read
Increase Llama 2's Latency and Throughput Performance by Up to 4X | Towards Data Science

Source: Towards Data Science

Real-world benchmarks for Llama-2 13B