Fix llava with multi-image inputs

intel-analytics / ipex-llm

Accelerate local LLM inference and finetuning (LLaMA, Mistral, ChatGLM, Qwen, Mixtral, Gemma, Phi, MiniCPM, Qwen-VL, MiniCPM-V, etc.) on Intel XPU (e.g., local PC with iGPU and NPU, discrete GPU such as Arc, Flex and Max); seamlessly integrate with llama.cpp, Ollama, HuggingFace, LangChain, LlamaIndex, vLLM, GraphRAG, DeepSpeed, Axolotl, etc

Apache License 2.0

6.75k stars 1.27k forks source link

Fix llava with multi-image inputs #12384

Closed Oscilloscope98 closed 1 week ago

Oscilloscope98 commented 1 week ago

Description

https://github.com/intel-analytics/ipex-llm/issues/12288

Fix llava with multi-image inputs, which cause x to be discontiguous.

PR validation: https://github.com/intel-analytics/ipex-llm-workflow/actions/runs/11794201827