This model was trained on a Deepseek V3.1 (reasoning) dataset with about 1K data entries. I will be uploading another Deepseek V3.1 distill version of Qwen3 8B, but fine-tuned with only 200x rows and benchmark them both to compare their performance. The benchmark results for this model will be published soon.

Downloads last month
250
GGUF
Model size
8B params
Architecture
qwen3
Hardware compatibility
Log In to view the estimation

4-bit

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

Dataset used to train Liontix/Qwen3-8B-Deepseek-V3.1-Distill-GGUF