Model Card for radm/Qwen2.5-32B-simpo-FP8

Model Details

Improved quality on hard tasks by 25 percent relative to the base model. Improved multilingual support.

Fine-tuning on A100 in 4-bit with unsloth using SIMPO and internal dataset

Eval results on ZebraLogic

Safetensors

Model size

32.8B params

Tensor type

BF16

F8_E4M3

Inference Providers NEW

This model is not currently available via any of the supported Inference Providers.

The model cannot be deployed to the HF Inference API: The model has no library tag.

Base model

Qwen/Qwen2.5-32B

Finetuned

Finetuned

(142)

this model