OpenReasoning-Nemotron-32B-AWQ

Method

Quantised using vllm-project/llm-compressor and the following configs:

recipe = GPTQModifier(targets="Linear", scheme="W4A16", ignore=["lm_head"])

Safetensors

Model size

5.7B params

Tensor type

I64

I32

BF16

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Base model

Qwen/Qwen2.5-32B

Finetuned

Finetuned

Quantized

(26)

this model