All decoder Linear layers and lm_head: GPTQ, symmetric INT4, group size 256, static actorder.
Calibration: HuggingFaceH4/ultrachat_200k, 512 samples, 1024 tokens, batch size 1.
No intentional mixed-precision language layers.

Full WikiText evaluation

EleutherAI/wikitext_document_level, wikitext-2-raw-v1, full run (2026-07-16).

Checkpoint	Word perplexity
`Qwen/Qwen3-1.7B`	23.086220
This GPTQ checkpoint	24.689014

Reproduce

python quantize.py --model-path /project/mlasw/share/huggingface/models--Qwen--Qwen3-1.7B --output-dir /path/to/output

Included: quantized safetensors, tokenizer/configuration, quantize.py, recipe.yaml, and versions.txt.

Validation and limitations

Saved scales were checked for finite values. Transformers load/generation smoke test is pending before release. Quantization quality can vary by language, prompt format, domain, and context length; validate the compiled deployment on target hardware.

simaai/Qwen3-1.7B-GPTQ-Safetensors

Qwen3-1.7B GPTQ INT4/G256 Safetensors

Quantization

Full WikiText evaluation

Reproduce

Validation and limitations

simaai/
Qwen3-1.7B-GPTQ-Safetensors