LLM EXPLORER 61,060 MODELS INDEXED

DSV4.1 Flash SAGE EXL3 3.30bpw by vcruz305

By vcruz305 · 529 downloads

DSV4.1 Flash SAGE EXL3 3.30bpw is an open-source language model by vcruz305. Features: 325.3b LLM, VRAM: 447.7GB, License: mit.

Base model:deepseek-ai/deepsee... Base model:quantized:deepseek-...   Deepseek   Deepseek v41   Exl3   Exllamav3   Mixed-precision   Moe   Quantization   Region:us   Safetensors   Sharded   Tensorflow

DSV4.1 Flash SAGE EXL3 3.30bpw Parameters and Internals

LLM NameDSV4.1 Flash SAGE EXL3 3.30bpw
Repository 🤗https://huggingface.co/vcruz305/DSV4.1-Flash-SAGE-EXL3-3.30bpw 
Base Model(s)  deepseek-ai/DeepSeek-V4.1-Flash   deepseek-ai/DeepSeek-V4.1-Flash
Model Size325.3b
Required VRAM447.7 GB
Updated2026-09-14
Maintainervcruz305
Model Typedeepseek_v41
Model Files  8.6 GB: 1-of-31   8.6 GB: 2-of-31   8.6 GB: 3-of-31   8.6 GB: 4-of-31   8.6 GB: 5-of-31   8.6 GB: 6-of-31   8.6 GB: 7-of-31   8.6 GB: 8-of-31   8.6 GB: 9-of-31   8.6 GB: 10-of-31   8.6 GB: 11-of-31   8.6 GB: 12-of-31   8.6 GB: 13-of-31   8.6 GB: 14-of-31   8.6 GB: 15-of-31   8.6 GB: 16-of-31   8.6 GB: 17-of-31   8.6 GB: 18-of-31   8.6 GB: 19-of-31   8.6 GB: 20-of-31   8.6 GB: 21-of-31   8.6 GB: 22-of-31   8.6 GB: 23-of-31   8.6 GB: 24-of-31   8.6 GB: 25-of-31   8.6 GB: 26-of-31   8.6 GB: 27-of-31   8.6 GB: 28-of-31   2.9 GB: 29-of-31   102.0 GB: 30-of-31   102.0 GB: 31-of-31
Model ArchitectureDeepseekV41ForCausalLM
Licensemit
Model Max Length1048576
Transformers Version5.6.0
Tokenizer ClassPreTrainedTokenizerFast
Beginning of Sentence Token<|begin▁of▁sentence|>
End of Sentence Token<|end▁of▁sentence|>