ipex-llm
ipex-llm copied to clipboard
intel-analytics
gemma3 and qwen3
I hope to add and adapt two useful models, gemma3 and qwen3, to the ollama function.
qwen3 works with the latest nightly build of ollma (ollama-ipex-llm-2.3.0b20250428-win.zip) on my Ultra 258V: https://github.com/ipex-llm/ipex-llm/releases/tag/v2.3.0-nightly
Hi @ybeetle8 , qwen3 model has been supported in our latest version. You may install it via pip install --pre --upgrade ipex-llm[cpp], see https://github.com/intel/ipex-llm/blob/main/docs/mddocs/Quickstart/ollama_quickstart.md for more usage details.
Where is “ollama-ipex-llm-2.3.0b20250428-ubuntu.tgz“”？
See https://github.com/intel/ipex-llm/blob/main/docs/mddocs/Quickstart/ollama_portable_zip_quickstart.md
qwen3适用于我的 Ultra 258V 上的最新 Nightly Build 的 ollma （ollama-ipex-llm-2.3.0b20250428-win.zip）：https://github.com/ipex-llm/ipex-llm/releases/tag/v2.3.0-nightly
Is Qwen3 really supported?
qwen3适用于我的 Ultra 258V 上的最新 Nightly Build 的 ollma （ollama-ipex-llm-2.3.0b20250428-win.zip）：https://github.com/ipex-llm/ipex-llm/releases/tag/v2.3.0-nightly
No, it reports the error "Error: llama runner process has terminated: exit status 1"
I tried Qwen3 and it can provide me an answer but only once. If I ask a second question, nothing happens - I don't see anything and have to restart.
On the latest nightly build for Arc 750, all models run slower than on the CPU, despite the GPU load being at 100%. Is it only like this on my machine? Windows one I mean, latest driver.
Hi all, gemma3 ollama is supported from ipex-llm[cpp]==2.3.0b20250529 .
You could try it again with pip install ipex-llm[cpp]==2.3.0b20250529.
Confirmed that both qwen3 and gemma3 models work with https://github.com/ipex-llm/ipex-llm/releases/download/v2.3.0-nightly/ollama-ipex-llm-2.3.0b20250612-ubuntu.tgz.
