LoganResearch
/

Qwen-32-B-4-Bit-Cygnus-Base

4-bit precision

Model card Files Files and versions

Qwen2.5-32B-Instruct - 4-bit (CYGNUS Base)

Qwen/Qwen2.5-32B-Instruct quantized to 4-bit (bitsandbytes NF4, double-quant, bfloat16 compute). This is the base model used by CYGNUS 1.0.

Original model (c) Qwen team, released under Apache-2.0.

Downloads last month: 142

Safetensors

Model size

34B params

Tensor type

F32

·

BF16

·

U8

·

Inference Providers NEW

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

Model tree for LoganResearch/Qwen-32-B-4-Bit-Cygnus-Base

Base model

Qwen/Qwen2.5-32B

Finetuned

Qwen/Qwen2.5-32B-Instruct

Quantized

(157)

this model