LLM EXPLORER 66,143 MODELS INDEXED

DeepSeek V4.1 Flash 1017 Sushi STQ 1.75bpw by beamster

By beamster · 605 downloads

DeepSeek V4.1 Flash 1017 Sushi STQ 1.75bpw is an open-source language model by beamster. Features: 261.3b LLM, VRAM: 204GB, License: mit, LLM Explorer Score: 0.35.

  Apple-silicon Base model:deepseek-ai/deepsee... Base model:quantized:deepseek-...   Deepseek v41   Fp8   Image-text-to-text   Moe   Region:us   Safetensors   Sharded   Stq   Sushi   Tensorflow

DeepSeek V4.1 Flash 1017 Sushi STQ 1.75bpw Parameters and Internals

LLM NameDeepSeek V4.1 Flash 1017 Sushi STQ 1.75bpw
Repository πŸ€—https://huggingface.co/beamster/DeepSeek-V4.1-Flash-1017-Sushi-STQ-1.75bpw 
Base Model(s)  deepseek-ai/DeepSeek-V4.1-Flash-1017   deepseek-ai/DeepSeek-V4.1-Flash-1017
Model Size261.3b
Required VRAM204 GB
Updated2026-10-30
Maintainerbeamster
Model Typedeepseek_v41
Model Files  102.0 GB: 47-of-48   102.0 GB: 48-of-48   1.8 GB   2.6 GB   2.6 GB   2.6 GB   2.7 GB   2.7 GB   2.7 GB   2.8 GB   2.8 GB   2.7 GB   2.8 GB   2.7 GB   2.8 GB   2.9 GB   2.8 GB   2.8 GB   3.7 GB   2.9 GB   3.2 GB   3.1 GB   3.2 GB   2.7 GB   2.8 GB   2.9 GB   3.0 GB   3.1 GB   3.2 GB   3.2 GB   3.4 GB   3.1 GB   3.2 GB   3.4 GB   3.1 GB   3.5 GB   3.4 GB   3.2 GB   3.5 GB   2.8 GB   3.0 GB   3.0 GB   3.0 GB
Model ArchitectureDeepseekV41ForCausalLM
Licensemit
Transformers Version5.6.0