search

janhq / cortex.tensorrt-llm

Cortex.Tensorrt-LLM is a C++ inference library that can be loaded by any server at runtime. It submodules NVIDIA’s TensorRT-LLM for GPU accelerated inference on NVIDIA's GPUs.
https://cortex.jan.ai/docs/cortex-tensorrt-llm
Apache License 2.0
40 stars 2 forks source link

fix: support Mistral v0.3 #52

Closed vansangpfiev closed 4 months ago

vansangpfiev commented 4 months ago

issue: https://github.com/janhq/cortex.tensorrt-llm/issues/50