LLM EXPLORER 66,613 MODELS INDEXED

LFM3 1.7B MLX OQ6e by UraionLabs

By UraionLabs · 267 downloads

LFM3 1.7B MLX OQ6e is an open-source language model by UraionLabs. Features: 1.7b LLM, VRAM: 1.4GB, Context: 256K, License: other.

  256k-context   6-bit   Agent   Agentic   Apple-silicon   Ar   Base model:liquidai/lfm3-1.7b Base model:quantized:liquidai/...   Conversational   Custom code   De   Edge   Edge-ai   En   Es   Fr   Function-calling   Hermes-agent   Hi   Hybrid-architecture   Id   It   Ja   Ko   Lfm   Lfm3   Lfm3-1.7b   Liquid   Liquid-ai   Local-inference   Local-llm   Long-context   Mac   Macos   Metal   Mixed-precision   Mlx   Mlx-vlm   Multilingual   Omlx   On-device   Openclaw   Oq   Oq6e   Oqe   Pl   Pt   Quantization   Quantized   Reasoning   Region:us   Ru   Safetensors   Short-convolution   Structured-output   Th   Thinking   Tool-calling   Tri-hybrid   Uraion-labs   Uraionlabs   Vi   Zh

LFM3 1.7B MLX OQ6e Parameters and Internals

LLM NameLFM3 1.7B MLX OQ6e
Repository πŸ€—https://huggingface.co/UraionLabs/LFM3-1.7B-MLX-oQ6e 
Base Model(s)  LiquidAI/LFM3-1.7B   LiquidAI/LFM3-1.7B
Model Size1.7b
Required VRAM1.4 GB
Updated2026-11-04
MaintainerUraionLabs
Model Typelfm3
Model Files  1.4 GB
Supported Languagesen ar zh fr de it ja ko pt es vi th id hi ru pl
Model ArchitectureLfm3ForCausalLM
Licenseother
Context Length262144
Model Max Length262144
Transformers Version5.12.1
Tokenizer ClassTokenizersBackend
Padding Token<|pad|>
Vocabulary Size128000

Best Alternatives to LFM3 1.7B MLX OQ6e

Best Alternatives
Context / RAM
Downloads
Likes
LFM3 1.7B256K / 3.5 GB32933624
LFM3 1.7B Base256K / 3.5 GB315293
LFM3 1.7B MLX Bf16256K / 3.5 GB138511
LFM3 1.7B NVFP4256K / 1.8 GB136016
LFM3 1.7B ONNX256K /  GB93213
LFM3 1.7B FP8256K / 2.2 GB89312
Cosmic Lightning256K / 3.5 GB1689
...3 1.7B Claude Code Safetensors256K / 3.5 GB9751
Acaua Lite256K / 3.5 GB1602
LFM3 1.7B Web ONNX256K /  GB5061
Note: green Score (e.g. "73.2") means that the model is better than UraionLabs/LFM3-1.7B-MLX-oQ6e.