LFM2.5-230M-MLX-OptiQ-8bit
LiquidAI/LFM2.5-230M, quantized with OptiQ by code-and-canvas for Apple Silicon.
communityOn disk0.2 GB
Memoryabout 2 GB
Use it
Install
pip install mlx-optiqServe it
An OpenAI- and Anthropic-compatible server on this Mac. Every app below can use it.
optiq serve --model code-and-canvas/LFM2.5-230M-MLX-OptiQ-8bitOptiQ Code
The terminal coding agent uses whatever model optiq serve is running.
optiq serve --model code-and-canvas/LFM2.5-230M-MLX-OptiQ-8bit
optiq codeOptiQ Lab
optiq lab --model code-and-canvas/LFM2.5-230M-MLX-OptiQ-8bitClaude Code
optiq run claude -m code-and-canvas/LFM2.5-230M-MLX-OptiQ-8bitCodex
optiq run codex -m code-and-canvas/LFM2.5-230M-MLX-OptiQ-8bitOpenCode
optiq run opencode -m code-and-canvas/LFM2.5-230M-MLX-OptiQ-8bitMistral Vibe
optiq run vibe -m code-and-canvas/LFM2.5-230M-MLX-OptiQ-8bitHermes Agent
optiq run hermes -m code-and-canvas/LFM2.5-230M-MLX-OptiQ-8bitOpenClaw
optiq run openclaw -m code-and-canvas/LFM2.5-230M-MLX-OptiQ-8bitoptiq run starts optiq serve if it is not running, sets the agent up for this session only, and stops the server when the agent exits.
mlx-lm, text only
from mlx_lm import load, generate
model, tokenizer = load("code-and-canvas/LFM2.5-230M-MLX-OptiQ-8bit")
print(generate(model, tokenizer, prompt="Hello", max_tokens=256))Details
| Family | LFM2.5 · LFM2.5 docs |
|---|---|
| Architecture | LFM2 |
| Base model | LiquidAI/LFM2.5-230M |
| Size on disk | 0.2 GB |
| Memory needed | about 2 GB |
| Images | no, text only |
| Downloads | 47 in the last month |
| Updated | 2026-06-27 |
| Hugging Face | code-and-canvas/LFM2.5-230M-MLX-OptiQ-8bit |