LLM EXPLORER 55,709 MODELS INDEXED

Fara1.5 27B OptiQ 4bit by mlx-community

By mlx-community · 256 downloads
HF Repo ↗ Copy URL

Fara1.5 27B OptiQ 4bit is an open-source language model by mlx-community. Features: 27b LLM, VRAM: 20.4GB, License: mit, Quantized.

  4-bit   4bit   Agent Base model:microsoft/fara1.5-2... Base model:quantized:microsoft...   Computer-use   Conversational   En   Image-text-to-text   Mlx   Optiq   Quantized   Qwen3 5   Region:us   Safetensors   Sharded   Tensorflow   Web-agent
Model Card on HF 🤗: https://huggingface.co/mlx-community/Fara1.5-27B-OptiQ-4bit 

Fara1.5 27B OptiQ 4bit Parameters and Internals

LLM NameFara1.5 27B OptiQ 4bit
Repository 🤗https://huggingface.co/mlx-community/Fara1.5-27B-OptiQ-4bit 
Base Model(s)  microsoft/Fara1.5-27B   microsoft/Fara1.5-27B
Model Size27b
Required VRAM20.4 GB
Updated2026-08-08
Maintainermlx-community
Model Typeqwen3_5
Model Files  5.3 GB: 1-of-4   5.4 GB: 2-of-4   5.4 GB: 3-of-4   4.3 GB: 4-of-4
Supported Languagesen
Quantization Type4bit
Model ArchitectureQwen3_5ForConditionalGeneration
Licensemit
Model Max Length262144
Transformers Version5.14.1
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Errorsreplace

Best Alternatives to Fara1.5 27B OptiQ 4bit

Best Alternatives
Context / RAM
Downloads
Likes
Qwen3.6 27B OptiQ 4bit256K / 18.7 GB1306559
Qwen3.5 27B OptiQ 4bit0K / 18.6 GB72693
Qwen3.6 27B UD Q3 K XL Mlx0K / 18.7 GB3953
Apax Q4 K M0K / 55.2 GB120
MLX Qwopus3.5 27B V3 6bit0K / 21.8 GB1413
... Opus Reasoning Distilled 4bit0K / 15.1 GB1308
Qwen3.5 27B 4bit DWQ0K / 15.2 GB596
Qwopus3.6 27B Coder 4bit0K / 16 GB3121
Qwen3.6 27B 4bit0K / 16 GB4367249
Qwopus3.5 27B V3 4bit0K / 16 GB231
Note: green Score (e.g. "73.2") means that the model is better than mlx-community/Fara1.5-27B-OptiQ-4bit.