# clm_2592F1F6FA15

subject: Hugging Face (/record/hugging-face.md)  
predicate: feature_added  
value: AutoGPTQ  
valid_from: 2023-08-23  
valid_to: null  
published: 2023-08-23  
recorded: 2026-11-10  
confidence: 0.70  
status: valid  
assertable: true  
superseded_by: null  
last_updated: 2026-11-10  
cite: AsOf. "Hugging Face feature added: AutoGPTQ", as of 2023-08-23. Claim clm_2592F1F6FA15, https://asof.is/claim/clm_2592F1F6FA15, accessed 2026-11-11. Source: https://huggingface.co/blog/gptq-integration.  
checked: 2026-11-10 · its source page states the value and the date  

> In the same spirit as our bitsandbytes collaboration , we have just integrated the AutoGPTQ library in Transformers, making it possible for users to quantize and run models in 8, 4, 3, or even 2-bit precision using the GPTQ algorithm

What https://huggingface.co/blog/gptq-integration says, word for word.

## Sources

- https://huggingface.co/blog/gptq-integration (primary, Hugging Face, published 2023-08-23)

---

AsOf is a temporal fact ledger: a structured, dated, sourced record of what is true and what changed, built for AI retrieval.