LLM EXPLORER 62,323 MODELS INDEXED

Swift 1.5 Qwen3.8 27B W4A16 AutoRound by ukisai

By ukisai · 229 downloads
HF Repo ↗ Copy URL

Swift 1.5 Qwen3.8 27B W4A16 AutoRound is an open-source language model by ukisai. Features: 27b LLM, VRAM: 18.7GB, License: other, LLM Explorer Score: 0.32.

  4-bit   Auto-round Base model:quantized:ukisai/sw... Base model:ukisai/swift-1.5-qw...   Conversational   Efficient-thinking   Endpoints compatible   Image-text-to-text   Post-training   Quantized   Qwen3 5   Qwen3 8   Reasoning   Region:us   Safetensors   Sharded   Tensorflow   Terminal-bench   Token-efficient
Model Card on HF 🤗: https://huggingface.co/ukisai/Swift-1.5-Qwen3.8-27b-W4A16-AutoRound 

Swift 1.5 Qwen3.8 27B W4A16 AutoRound Parameters and Internals

LLM NameSwift 1.5 Qwen3.8 27B W4A16 AutoRound
Repository 🤗https://huggingface.co/ukisai/Swift-1.5-Qwen3.8-27b-W4A16-AutoRound 
Base Model(s)  ukisai/Swift-1.5-Qwen3.8-27b   ukisai/Swift-1.5-Qwen3.8-27b
Model Size27b
Required VRAM18.7 GB
Updated2026-09-27
Maintainerukisai
Model Typeqwen3_5
Model Files  2.5 GB: 1-of-5   5.0 GB: 2-of-5   5.0 GB: 3-of-5   5.0 GB: 4-of-5   1.2 GB: 5-of-5   0.8 GB
Model ArchitectureQwen3_5ForConditionalGeneration
Licenseother
Model Max Length262144
Transformers Version5.14.1
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Errorsreplace

Best Alternatives to Swift 1.5 Qwen3.8 27B W4A16 AutoRound

Best Alternatives
Context / RAM
Downloads
Likes
...wopus3.8 27B Flash Mlx OQ4e 1M1024K / 16.9 GB2111
Agens Pilot1024K / 54.7 GB1742
Qwen3.6 27B0K / 55.4 GB67281382197
Qwen3.8 27B0K / 55.6 GB28753
Qwen3.8 27B FP80K / 0.4 GB123157416
Qwen3.8 27B0K / 55.6 GB272721
Qwen3.8 27B MLX0K / 16 GB141118
Qwen3.6 27B NVFP40K / 22 GB1887525419
Qwen3.5 27B0K / 55.2 GB25362941028
Qwen3.6 27B FP80K / 0.4 GB7740472342
Note: green Score (e.g. "73.2") means that the model is better than ukisai/Swift-1.5-Qwen3.8-27b-W4A16-AutoRound.