List view

Support KV cache quantization, especially for vLLM inference

high priority

Status: Open.

#920 In intel/auto-round;

· xin3he opened on Oct 20, 2025

Intel GPU takes much more memory than expected

Intel GPU

Status: Open.

#990 In intel/auto-round;

· xin3he opened on Nov 4, 2025

Refactor modelling replacement code

high priority

Status: Open.

#899 In intel/auto-round;

· yiliu30 opened on Oct 15, 2025

enable Qwen3-VL-MoE model quantization, save and vllm loading

high priority

Status: Open.

#942 In intel/auto-round;

· WeiweiZhang1 opened on Oct 24, 2025

enhance auto device map for moe model

Status: Open.

#976 In intel/auto-round;

· wenhuach21 opened on Oct 31, 2025

Card 0 OOM during block inference

Status: Open.

#988 In intel/auto-round;

· xin3he opened on Nov 4, 2025