DeepSeek-R1-Distill-Qwen-32B — GGUF (Q4_K_M)
bartowski's Q4_K_M GGUF for the reasoning-distilled 32B model, re-hosted with measured performance for the 1bit engine on Strix Halo.
Contents
DeepSeek-R1-Distill-Qwen-32B-Q4_K_M.gguf
Measured performance (Strix Halo, Vulkan)
pp512: 237 tok/s · tg128: 9.3 tok/s
Running it
1bit serve -m DeepSeek-R1-Distill-Qwen-32B-Q4_K_M.gguf --device vulkan
Attribution
- Model: deepseek-ai/DeepSeek-R1-Distill-Qwen-32B, MIT.
- Quantization: bartowski/DeepSeek-R1-Distill-Qwen-32B-GGUF.
- License: MIT, inherited from the base model.
- Downloads last month
- 95
Hardware compatibility
Log In to add your hardware
4-bit
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for 1bit-MONSTER/DeepSeek-R1-Distill-Qwen-32B-GGUF
Base model
deepseek-ai/DeepSeek-R1-Distill-Qwen-32B