LLM EXPLORER 67,711 MODELS INDEXED

Qwen3.8 Flash Next 3.48BPW by Beling86

By Beling86 · 128 downloads

Qwen3.8 Flash Next 3.48BPW is an open-source language model by Beling86. Features: 15.3b LLM, VRAM: 106.4GB, Context: 256K, LLM Explorer Score: 0.31.

  4-bit   Auto-round   Autoround Base model:quantized:qwen/qwen... Base model:qwen/qwen3.8-flash-...   Conversational   Endpoints compatible   Fp8   Int3   Mixed-precision   Moe   Qwen4 exp text   Region:us   Safetensors   Sharded   Tensorflow   Vllm

Qwen3.8 Flash Next 3.48BPW Parameters and Internals

LLM NameQwen3.8 Flash Next 3.48BPW
Repository πŸ€—https://huggingface.co/Beling86/Qwen3.8-Flash-Next-3.48BPW 
Base Model(s)  Qwen/Qwen3.8-Flash-Next   Qwen/Qwen3.8-Flash-Next
Model Size15.3b
Required VRAM106.4 GB
Updated2026-11-17
MaintainerBeling86
Model Typeqwen4_exp_text
Model Files  5.4 GB: 1-of-11   5.4 GB: 2-of-11   5.4 GB: 3-of-11   5.4 GB: 4-of-11   5.4 GB: 5-of-11   5.4 GB: 6-of-11   5.4 GB: 7-of-11   5.4 GB: 8-of-11   5.4 GB: 9-of-11   5.2 GB: 10-of-11   1.3 GB: 11-of-11   2.9 GB: 1-of-17   3.2 GB: 2-of-17   3.2 GB: 3-of-17   3.2 GB: 4-of-17   3.2 GB: 5-of-17   3.2 GB: 6-of-17   3.2 GB: 7-of-17   3.2 GB: 8-of-17   3.2 GB: 9-of-17   3.2 GB: 10-of-17   3.2 GB: 11-of-17   3.2 GB: 12-of-17   3.2 GB: 13-of-17   3.2 GB: 14-of-17   3.2 GB: 15-of-17   3.2 GB: 16-of-17   0.4 GB: 17-of-17
Model ArchitectureQwen4ExpForCausalLM
Context Length262144
Model Max Length262144
Transformers Version5.22.2
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Vocabulary Size248320
Errorsreplace