Support GPTQ/Marlin format quantization (4bit weight, f16 input) #89
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
GPTQ/Marlin quantization
Candle-vllm now supports GPTQ (Marlin kernel), you may supply the
quant
(marlin) anddtype
(f16) parameters if you haveMarlin
format quantized models, such as:Tested speed: 115 tokens/s (batch size = 1), 753 tokens/s (batch size=16) for LLaMa3.1 8B. (almost double performance for single query compared to bf16 format)
You may use
AutoGPTQ
to transform a model to marlin format by loading the (quantized) model and supply theuse_marlin=True
inAutoGPTQ
(which will generate marlin format quantized model once you callsave_pretrained
).Note: only 4bit GPTQ quantization supported for marlin format at the moment, and the input data type should be
f16
(--dtype f16). You need also renamed the transformed marlin format weight to "model.safetensors" and copy the "tokenizer.json" from the source model folder.Further plan: in-situ convertion of any quantized models to marlin format for speeding up inference.