Repository Issue Activity (beta feature)

#

#2568 Getting "KeyError: language_model.language_model.layers.0.mlp.experts.w2_weight_packed" while trying to infer AWQ quantized Qwen3.5-122B model using vllm

#2637 [Bug]: When I use the GPTQ algorithm to quantize a 4B model, the CPU memory usage is extremely high—approaching 200GB. Could someone explain to me why this is the case?

#

#

#

#

#

#

#2515 [Bug]: Fix ML3 FP8 documentation example

#2628 Asymmetric W4A16 AWQ quantization for MoE models fails at inference in vLLM

#

#

#

#

#2490 Consolidate Intermediate Offloading

#2624 Q2 2026 Roadmap

#

#2180 [RFC] [Performance Refactor][Distributed] Sequential Onloading with Data-Parallel Calibration and Weight-Parallel Optimization

#

#2056 [Bug]: InternVL-8B quantize failed with error: AttributeError: 'dict' object has no attribute 'last_hidden_state'

#

#

#

#