Dongjie Shi

Results 57 comments of Dongjie Shi

Arc770 IPEX-LLM 的性能问题

please try with latest Docker image: intelanalytics/ipex-llm-serving-xpu:2.2.0-SNAPSHOT as the data reviewed with you, the 1xARC/2xARC performance should be fine, 4xARC performance is degraded due to the high communication overhead, and...

Determining if AMX is in use by ollama

Sorry, we don't have any plan for optimizing ollama for Xeon/AMX yet.

MTL platform with ARC 770 cannot allocate memory block with size lager than 4GB when running vLLM Qwen2-VL-2B

We are validating 0.6.2 version, and Qwen2-VL model, will notify you once it's ready. Thanks.

Running vLLM service benchmark(1xARC770) with Qwen1.5-14B-Chat model failed(compression weight:SYM_INT4).

closed since no update for a long time

Support DeepSeek-Coder-v1.5 7B

fixed

vLLM 0.5.4 failure to start the TP+ PP mode on 8 ARC

fixed

IPEX-LLM 运行源2.0 M32量化版失败 on Intel ARC

fixed