4 Issues closed by 4 people

• Linux 下ipex-ollama 启动模型乱码

  #13219 closed Jun 16, 2025

• OCR slows down with ollama-vision models

  #13195 closed Jun 15, 2025

• ollama ipex-llm crashes when running tinyllama llama 3.1b and qwen 2.5

  #13221 closed Jun 14, 2025

• OOM issue when using --low-bit-model-path

  #13202 closed Jun 13, 2025

6 Issues opened by 6 people

• Training doesn't start QLoRA fine tune BMG

  #13226 opened Jun 18, 2025

• Crashes with DeepSeek-R1-0528-Qwen3-8B-UD-Q8_K_XL (maybe others)

  #13225 opened Jun 17, 2025

• qwen32B hung when running 20K/12K w/ 4 GPU

  #13224 opened Jun 17, 2025

• ollama ipex-llm crashes when running tinyllama llama 3.1b and qwen 2.5

  #13223 opened Jun 16, 2025

• Can the docker images be made smaller?

  #13222 opened Jun 15, 2025

• Ollama offload to the CPU then crash

  #13220 opened Jun 12, 2025

17 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• ollama-ipex fails running any models - A380 - llama runner process terminated

  #13212 commented on Jun 11, 2025 • 0 new comments

• multi-modal model chunked-prefill report error on ARC

  #13217 commented on Jun 12, 2025 • 0 new comments

• A770 can't run deepseek R1 Q4 with flashmoe

  #13183 commented on Jun 12, 2025 • 0 new comments

• 段错误 核心已转移

  #13218 commented on Jun 13, 2025 • 0 new comments

• Portable Builds Upstream Version Update Request

  #13184 commented on Jun 13, 2025 • 0 new comments

• Ollama run crashes using `ipex-llm-inference-cpp-xpu` docker image - A750 - Seg Fault

  #13213 commented on Jun 14, 2025 • 0 new comments

• gemma3 and qwen3

  #13123 commented on Jun 14, 2025 • 0 new comments

• How to get the supported models list

  #13200 commented on Jun 14, 2025 • 0 new comments

• ollama + deepseek v2: The number of work-items in each dimension of a work-group cannot exceed {512, 512, 512} for this device

  #12839 commented on Jun 16, 2025 • 0 new comments

• Ollama-ipex cannot support Gemma3 4B、Gemma3 12B、 phi4-mini 3.8B，but the official Ollama can.

  #13192 commented on Jun 16, 2025 • 0 new comments

• Qwen 2.5-vl support？

  #13188 commented on Jun 16, 2025 • 0 new comments

• 4xARC 770 on w5-3423 slow performance (half of what is reported here)

  #13214 commented on Jun 16, 2025 • 0 new comments

• Support for gemma3 from google

  #12963 commented on Jun 17, 2025 • 0 new comments

• The NPU version of llama.cpp did not return an appropriate response on Intel Core Ultra 7 268V

  #13074 commented on Jun 18, 2025 • 0 new comments

• B60 Performance Issue with INT4

  #13173 commented on Jun 18, 2025 • 0 new comments

• Update Dockerfile

  #13069 commented on Jun 15, 2025 • 0 new comments

• Add PyTorch 2.7 Support for Intel GPU

  #13120 commented on Jun 15, 2025 • 0 new comments