Junhong Liu

Results 21 comments of Junhong Liu
trafficstars

Error trying to build the visual encoder for llava-v1.6-34b-hf using build_visual_engine.py

@alexemme If you have no further questions, we will close it in a week.

Tensorrt_llm install issues

@KarihaZach Do you still have the question? If not, we will close it soon.

make -C docker run LOCAL_USER=1 FAILED

@LoverLost Do you still have the problem? If not, we will close it soon.

Fail to buid inference trt_llm image : make: *** [Makefile:64: release_build] Error 1

@dadaguai-jiangjun Could you try to use the latest version of TrtLLM?

kv_cache_reuse breaking on awq quantized model

@Bhuvanesh09 if you have no further questions, we will close this issue in one week.

result is different from 0.9.0 and 0.10.0，and speed has decreased when update version

@sundayKK, please try to use the latest version of TrtLLM.

Fix prompt_table_data empty tensor shape error

@BasicCoder Thank you for your interest in TrtLLm. We have fixed it in our internal version, and it will be updated on GitHub soon.

problem with output_log_probs

Could you tell me more details about your questions?

Converting the Large world model in TensorRT LLM

cc @AdamzNV @ncomly-nvidia @laikhtewari for vis.

Feature request: Ability to provide your own feature mask in ModelRunner class.

cc @AdamzNV @ncomly-nvidia @laikhtewari for vis.