Weights

Models, kept on Hugging Face.

Open Hugging Face ↗
DeepSeek V4.1 Flash 1017 · 4.0 bpwbertholomus/DeepSeek-V4.1-Flash-1017-EXL3-4.0bpw
BASEDeepSeek-V4.1-Flash-1017
FORMATEXL3
hf download bertholomus/DeepSeek-V4.1-Flash-1017-EXL3-4.0bpw
VIEW ON HUGGING FACE ↗
DeepSeek V4.1 Flash 1017 · 2.5 bpwbertholomus/DeepSeek-V4.1-Flash-1017-EXL3-2.5bpw
BASEDeepSeek-V4.1-Flash-1017
FORMATEXL3
hf download bertholomus/DeepSeek-V4.1-Flash-1017-EXL3-2.5bpw
VIEW ON HUGGING FACE ↗
GLM-5.3 · 3.0 bpwbertholomus/GLM-5.3-EXL3-3.0bpw
BASEzai-org/GLM-5.3
FORMATEXL3
hf download bertholomus/GLM-5.3-EXL3-3.0bpw
VIEW ON HUGGING FACE ↗
DeepSeek V4.1 Flash · TensorFold TP4bertholomus/DeepSeek-V4.1-Flash-TensorFold-TP4-4xGB10
BASEDeepSeek-V4.1-Flash-1017
FORMATTensorFold · 4× GB10
hf download bertholomus/DeepSeek-V4.1-Flash-TensorFold-TP4-4xGB10
VIEW ON HUGGING FACE ↗
DeepSeek V4.1 Flash · TensorFold TP2bertholomus/DeepSeek-V4.1-Flash-TensorFold-TP2-2xGB10
BASEDeepSeek-V4.1-Flash-1017
FORMATTensorFold · 2× GB10
hf download bertholomus/DeepSeek-V4.1-Flash-TensorFold-TP2-2xGB10
VIEW ON HUGGING FACE ↗
DeepSeek V4.1 Flash · DSpark TP4bertholomus/DeepSeek-V4.1-Flash-DSpark-TP4-4xGB10-Recipe
BASEDeepSeek-V4.1-Flash
FORMATSGLang · 4× GB10
hf download bertholomus/DeepSeek-V4.1-Flash-DSpark-TP4-4xGB10-Recipe
VIEW ON HUGGING FACE ↗
GLM-5.3 Flash · NVFP4 TP4bertholomus/GLM-5.3-Flash-NVFP4-4xGB10-TP4
BASERedHatAI/GLM-5.3-Flash-NVFP4
FORMATvLLM · 4× GB10
hf download bertholomus/GLM-5.3-Flash-NVFP4-4xGB10-TP4
VIEW ON HUGGING FACE ↗
DeepSeek V4 Flash 0731 · Graph-8, four nodesbertholomus/DeepSeek-V4-Flash-0731-DSpark-Graph8-4xGB10
BASEDeepSeek-V4-Flash-0731
FORMATvLLM · 4× GB10
hf download bertholomus/DeepSeek-V4-Flash-0731-DSpark-Graph8-4xGB10
VIEW ON HUGGING FACE ↗
DeepSeek V4 Flash 0731 · Graph-8, two nodesbertholomus/DeepSeek-V4-Flash-0731-DSpark-Graph8
BASEDeepSeek-V4-Flash-0731
FORMATvLLM · 2× GB10
hf download bertholomus/DeepSeek-V4-Flash-0731-DSpark-Graph8
VIEW ON HUGGING FACE ↗