LLM EXPLORER 62,751 MODELS INDEXED

Swift 1.5 Qwen3.8 Flash Next W4A16 Attn8 FP8PLE by todiadiyatmo

By todiadiyatmo · 54 downloads
HF Repo ↗ Copy URL

Swift 1.5 Qwen3.8 Flash Next W4A16 Attn8 FP8PLE is an open-source language model by todiadiyatmo. Features: 71.1b LLM, VRAM: 55.9GB, License: other, LLM Explorer Score: 0.28.

  4-bit   Auto-round   Autoround Base model:quantized:ukisai/sw... Base model:ukisai/swift-1.5-qw...   Conversational   Fp8   Image-text-to-text   Int4   Int8   Mixed-precision   Mtp   Qwen3.8   Qwen4 exp   Region:us   Safetensors   Sharded   Swift   Tensorflow   Vllm
Model Card on HF 🤗: https://huggingface.co/todiadiyatmo/Swift-1.5-Qwen3.8-Flash-Next-W4A16-Attn8-FP8PLE 

Swift 1.5 Qwen3.8 Flash Next W4A16 Attn8 FP8PLE Parameters and Internals

LLM NameSwift 1.5 Qwen3.8 Flash Next W4A16 Attn8 FP8PLE
Repository 🤗https://huggingface.co/todiadiyatmo/Swift-1.5-Qwen3.8-Flash-Next-W4A16-Attn8-FP8PLE 
Base Model(s)  ukisai/Swift-1.5-Qwen3.8-Flash-Next-W4A16-AutoRound   ukisai/Swift-1.5-Qwen3.8-Flash-Next-W4A16-AutoRound
Model Size71.1b
Required VRAM55.9 GB
Updated2026-10-02
Maintainertodiadiyatmo
Model Typeqwen4_exp
Model Files  2.7 GB: 1-of-1   0.1 GB   1.4 GB: 1-of-81   1.4 GB: 35-of-81   1.4 GB: 36-of-81   1.4 GB: 37-of-81   1.4 GB: 38-of-81   1.4 GB: 39-of-81   1.4 GB: 40-of-81   1.4 GB: 41-of-81   1.4 GB: 42-of-81   1.4 GB: 43-of-81   1.4 GB: 44-of-81   1.4 GB: 45-of-81   1.4 GB: 46-of-81   1.4 GB: 47-of-81   1.4 GB: 48-of-81   1.4 GB: 49-of-81   1.4 GB: 50-of-81   1.4 GB: 51-of-81   1.4 GB: 52-of-81   1.4 GB: 53-of-81   1.4 GB: 54-of-81   1.4 GB: 55-of-81   1.4 GB: 56-of-81   1.4 GB: 57-of-81   1.4 GB: 58-of-81   1.4 GB: 59-of-81   1.4 GB: 60-of-81   1.4 GB: 61-of-81   1.4 GB: 62-of-81   1.4 GB: 63-of-81   1.4 GB: 64-of-81   1.4 GB: 65-of-81   1.4 GB: 66-of-81   1.4 GB: 67-of-81   1.4 GB: 68-of-81   1.4 GB: 69-of-81   1.4 GB: 70-of-81   1.4 GB: 71-of-81
Model ArchitectureQwen4ExpForConditionalGeneration
Licenseother
Transformers Version5.16.1