LLM EXPLORER 66,291 MODELS INDEXED

Ukisai Swift 1.5 Qwen3.8 27B Exl3 3.50bpw H4 Vast Tmp by Starw1

By Starw1 · 44 downloads
HF Repo ↗ Copy URL

Ukisai Swift 1.5 Qwen3.8 27B Exl3 3.50bpw H4 Vast Tmp is an open-source language model by Starw1. Features: 27b LLM, VRAM: 15GB, License: other, LLM Explorer Score: 0.29.

Base model:finetune:qwen/qwen3...   Base model:qwen/qwen3.8-27b   Conversational   Efficient-thinking   Endpoints compatible   Exl3   Image-text-to-text   Post-training   Qwen3 5   Qwen3 8   Reasoning   Region:us   Safetensors   Sharded   Tensorflow   Terminal-bench   Token-efficient
Model Card on HF 🤗: https://huggingface.co/Starw1/ukisai-Swift-1.5-Qwen3.8-27B-exl3-3.50bpw-H4-vast-tmp 

Ukisai Swift 1.5 Qwen3.8 27B Exl3 3.50bpw H4 Vast Tmp Parameters and Internals

LLM NameUkisai Swift 1.5 Qwen3.8 27B Exl3 3.50bpw H4 Vast Tmp
Repository 🤗https://huggingface.co/Starw1/ukisai-Swift-1.5-Qwen3.8-27B-exl3-3.50bpw-H4-vast-tmp 
Base Model(s)  Qwen/Qwen3.8-27B   Qwen/Qwen3.8-27B
Model Size27b
Required VRAM15 GB
Updated2026-10-31
MaintainerStarw1
Model Typeqwen3_5
Model Files  8.5 GB: 1-of-2   6.5 GB: 2-of-2
Model ArchitectureQwen3_5ForConditionalGeneration
Licenseother
Model Max Length262144
Transformers Version5.8.0.dev0
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Errorsreplace

Best Alternatives to Ukisai Swift 1.5 Qwen3.8 27B Exl3 3.50bpw H4 Vast Tmp

Best Alternatives
Context / RAM
Downloads
Likes
...wopus3.8 27B Flash Mlx OQ4e 1M1024K / 16.9 GB2111
Agens Pilot1024K / 54.7 GB863
Qwen3.8 27B NPU2256K /  GB2524
JEV 27B VL0K / 55.6 GB1525286966
Clef0K / 54.7 GB220432237
Qwen3.8 27B NVFP4 DGX SPARK V20K / 19.8 GB11317113
Qwen3.6 27B0K / 55.4 GB67281382197
Qwen3.8 27B0K / 55.6 GB28753
Qwen3.8 27B0K / 55.6 GB272721
Qwen3.8 27B MLX0K / 16 GB141118
Note: green Score (e.g. "73.2") means that the model is better than Starw1/ukisai-Swift-1.5-Qwen3.8-27B-exl3-3.50bpw-H4-vast-tmp.