LLM EXPLORER 66,845 MODELS INDEXED

Mavery Geselle 2B MLX 4bit by rhymeas

By rhymeas · 77 downloads
HF Repo ↗ Copy URL

Mavery Geselle 2B MLX 4bit is an open-source language model by rhymeas. Features: 2b LLM, VRAM: 1.7GB, License: apache-2.0, Quantized, LLM Explorer Score: 0.26.

  4-bit   4bit   Apple-silicon Base model:quantized:qwen/qwen...   Base model:qwen/qwen3.5-2b   Conversational   De   Dictation   En   Mlx   Punctuation   Quantized   Qwen3 5   Region:us   Safetensors   Text-cleanup   Text-normalization   Truecasing
Model Card on HF 🤗: https://huggingface.co/rhymeas/Mavery-Geselle-2B-MLX-4bit 

Mavery Geselle 2B MLX 4bit Parameters and Internals

LLM NameMavery Geselle 2B MLX 4bit
Repository 🤗https://huggingface.co/rhymeas/Mavery-Geselle-2B-MLX-4bit 
Base Model(s)  Qwen/Qwen3.5-2B   Qwen/Qwen3.5-2B
Model Size2b
Required VRAM1.7 GB
Updated2026-09-20
Maintainerrhymeas
Model Typeqwen3_5
Model Files  1.7 GB
Supported Languagesde en
Quantization Type4bit
Model ArchitectureQwen3_5ForConditionalGeneration
Licenseapache-2.0
Model Max Length262144
Transformers Version4.57.0.dev0
Tokenizer ClassTokenizersBackend
Padding Token<|endoftext|>
Errorsreplace

Best Alternatives to Mavery Geselle 2B MLX 4bit

Best Alternatives
Context / RAM
Downloads
Likes
Qwen3.5 2B M40 6bit0K / 1.5 GB4631
Qwen3.5 2B Base 6bit0K / 1.5 GB4171
...en35 2B Base Cotypist Mlx 4bit0K / 1.1 GB1001
Qwen3.5 2B 4bit Slim0K / 1.1 GB2271
...Bonsai 2 2B ONNX WebGPU Vision0K /  GB1881
Synapse Alpha 1.0 4bit0K / 1.1 GB431
Qwen3.5 2B MLX 4bit0K / 1.7 GB123102
Loopl 2B 4bit0K / 1.7 GB1371
Qwen3.5 2B 6bit0K / 2.2 GB86807
Qwen3.8 2B Distill Mlx 5Bit0K / 1.3 GB3871
Note: green Score (e.g. "73.2") means that the model is better than rhymeas/Mavery-Geselle-2B-MLX-4bit.