text-generation-inference/backends/v3
Daniël de Kok c7b495f97d hotfix: avoid non-prefilled block use when using prefix caching (#2489)
The minimum batch size logic could cause prefix blocks to be
deallocated without prefill. The next allocation of the same
prefix would then use garbage blocks.
2024-09-25 06:13:11 +00:00
..
benches Keeping the benchmark somewhere (#2401) 2024-09-25 06:05:43 +00:00
src hotfix: avoid non-prefilled block use when using prefix caching (#2489) 2024-09-25 06:13:11 +00:00
build.rs Rebase TRT-llm (#2331) 2024-09-25 05:55:39 +00:00
Cargo.toml fix: bump minijinja version and add test for llama 3.1 tools (#2463) 2024-09-25 06:11:21 +00:00