Issues

• Maintaining Weight Equivalence

  OpenGVLab/OmniQuant#115 · Filter by author kimjoohyungsd opened on Jul 1, 2026

• Question about Activation quantization

  OpenGVLab/OmniQuant#113 · Filter by author Tfloow opened on Apr 11, 2026

• Question regarding Figure A5

  OpenGVLab/OmniQuant#112 · Filter by author ghost opened on Feb 2, 2026

• Bad performance with Auto-GPTQ kernel.d

  OpenGVLab/OmniQuant#111 · Filter by author jiuyixu25 opened on Jan 16, 2026

• Can omniquant be used on CNN models, similar to YOLO?

  OpenGVLab/OmniQuant#109 · Filter by author yz302 opened on Dec 16, 2025

• RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:7 and cuda:0! (when checking argument for argument index in method wrapper_CUDA__index_select)

  OpenGVLab/OmniQuant#105 · Filter by author forcekkk opened on Apr 11, 2025

• I encounter a error: "AttributeError: 'LlamaAttention' object has no attribute 'rotary_emb'",when i run code with llama-1-7b. It happened in int_llama_layer.py: self.rotary_emb = copy.deepcopy(org_module.rotary_emb)

  OpenGVLab/OmniQuant#104 · Filter by author WX-yh opened on Jan 5, 2025
  3

• I encounter a error: "AttributeError: 'LlamaAttention' object has no attribute 'rotary_emb'",when i run code with llama-1-7b.

  OpenGVLab/OmniQuant#103 · Filter by author WX-yh opened on Jan 5, 2025
  1

• Fail to reproduce the result of w2a16 using llama2 7b

  OpenGVLab/OmniQuant#102 · Filter by author stackByStack opened on Dec 16, 2024
  1

• Obtained different PPL for Wikitext and C4 compared to results reported in the paper

  OpenGVLab/OmniQuant#95 · Filter by author yc2367 opened on Sep 12, 2024
  2

• Error when evaluating MMLU

  OpenGVLab/OmniQuant#91 · Filter by author zjq0455 opened on Sep 4, 2024
  2

• how to enable llama3-8b int4 awq models

  OpenGVLab/OmniQuant#90 · Filter by author FlexLaughing opened on Aug 29, 2024

• RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cpu and cuda:0! (when checking argument for argument mat2 in method wrapper_CUDA_bmm)

  OpenGVLab/OmniQuant#89 · Filter by author mcpaulgeorge opened on Aug 5, 2024
  5

• The version of transformers, auto_gptq, autoawq

  OpenGVLab/OmniQuant#88 · Filter by author zhangfzR opened on Aug 5, 2024

• Quantify tinyllama-1.1B-Chat-v1.0, a CUDA assertion error occurred

  OpenGVLab/OmniQuant#87 · Filter by author causeof-you opened on Jul 12, 2024

• [New Feature] Seek MLA Supported by Smooth

  OpenGVLab/OmniQuant#86 · Filter by author RanchiZhao opened on Jul 10, 2024

• question about let

  OpenGVLab/OmniQuant#85 · Filter by author mxjmtxrm opened on Jul 8, 2024

• [Model Request] MiniCPM

  OpenGVLab/OmniQuant#84 · Filter by author RanchiZhao opened on Jul 2, 2024

• Questions about quantization

  OpenGVLab/OmniQuant#81 · Filter by author mxjmtxrm opened on Jun 10, 2024
  1

• Which bug do you fix for auto_gptq

  OpenGVLab/OmniQuant#79 · Filter by author BaohaoLiao opened on Jun 2, 2024
  1

• Questions regarding Infusing Omniquant into MLC

  OpenGVLab/OmniQuant#77 · Filter by author BuildBackBuehler opened on Apr 29, 2024
  3

• OPT-30B

  OpenGVLab/OmniQuant#76 · Filter by author Arthur-Ling opened on Apr 26, 2024

• Llama-3-8B

  OpenGVLab/OmniQuant#75 · Filter by author hsb1995 opened on Apr 22, 2024
  5

• TypeError: FalconRotaryEmbedding.forward() missing 1 required positional argument: position_ids

  OpenGVLab/OmniQuant#72 · Filter by author luchangli03 opened on Apr 11, 2024

• Other Task

  OpenGVLab/OmniQuant#67 · Filter by author hsb1995 opened on Mar 20, 2024
  1