huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-09-10 20:04:52 +00:00
Code Issues Packages Projects Releases Wiki Activity

Iterated on Pedro's comments

Browse Source
This commit is contained in:
Merve Noyan 2023-09-08 13:01:58 +02:00 committed by GitHub
parent e82259106c
commit 704cd18402
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
docs/source/conceptual/quantization.md
View File

@ -4,7 +4,7 @@ TGI offers GPTQ and bits-and-bytes quantization to quantize large language model
## Quantization with GPTQ
GPTQ is a post-training quantization method to make the model smaller. It quantizes each weight by finding a compressed version of that weight, that will yield a minimum mean squared error like below 👇
GPTQ is a post-training quantization method to make the model smaller. It quantizes the layers by finding a compressed version of that weight, that will yield a minimum mean squared error like below 👇
Given a layer \\(l\\) with weight matrix \\(W_{l}\\) and layer input \\(X_{l}\\), find quantized weight \\(\\hat{W}_{l}\\):
@ -17,7 +17,7 @@ TGI allows you to both run an already GPTQ quantized model (see available models
docker run --gpus all --shm-size 1g -p 8080:80 -v $volume:/data ghcr.io/huggingface/text-generation-inference:latest --model-id $model --quantize gptq
```
Note that TGI's GPTQ implementation is different than [AutoGPTQ](https://github.com/PanQiWei/AutoGPTQ).
Note that TGI's GPTQ implementation doesn't use [AutoGPTQ](https://github.com/PanQiWei/AutoGPTQ) under the hood. However, models quantized using AutoGPTQ or Optimum can still be served by TGI.
To quantize a given model using GPTQ with a calibration dataset, simply run