Search results

Support DeepSeekV3-style block FP8 quantization with CT

deepseekRelated to DeepSeek modelsRelated to DeepSeek models

needs-rebase

quantization

Status: Open (in progress).

vllm-project/vllmnumber 20279

#20279 In vllm-project/vllm;

· mgoin opened on Jun 30, 2025

[Kernel] Add ExLlamaV2 Weight Quantization Support

ci/build

kernel

llamaRelated to Llama modelsRelated to Llama models

needs-rebase

quantization

Status: Draft (not ready).

vllm-project/vllmnumber 11348

#11348 In vllm-project/vllm;

· AlpinDale opened on Dec 19, 2024