LFM2.5-1.2B-Instruct-OptiQ-4bit

mlx-community/LFM2.5-1.2B-Instruct-OptiQ-4bit

Built with mlx-optiq, the MLX-native toolkit to quantize, fine-tune, and serve LLMs locally on Apple Silicon, no PyTorch and no cloud. All OptiQ quants · Docs

An OptiQ mixed-precision quant of LiquidAI/LFM2.5-1.2B-Instruct. 825 MB on disk, down from 2.34 GB at bf16, and it scores 79% on IFEval and 70% on GSM8K at that size.

LFM2.5 is a hybrid architecture: convolutional blocks interleaved with full attention, so only a few blocks carry a KV cache. OptiQ measures each layer's sensitivity and assigns per-layer bit-widths, keeping 8-bit where it matters and 4-bit elsewhere.

What it is

Capability Score

Six-metric mean, the standard OptiQ eval.

Instruction-following and grade-school maths are the strengths here, and it writes usable Python at nearly 50% pass@1 for a model under a gigabyte. Long-context multi-hop retrieval is the weak spot.

Run it

pip install mlx-optiq

from mlx_lm import load, generate

model, tok = load("mlx-community/LFM2.5-1.2B-Instruct-OptiQ-4bit")
prompt = tok.apply_chat_template(
    [{"role": "user", "content": "Write a haiku about compilers."}],
    tokenize=False, add_generation_prompt=True,
)
print(generate(model, tok, prompt=prompt, max_tokens=300))

Serve it with the bundled mixed-precision KV cache:

optiq serve --model mlx-community/LFM2.5-1.2B-Instruct-OptiQ-4bit --kv-config kv_config.json

Tool calling

LFM2.5 writes Pythonic calls between special tokens:

<|tool_call_start|>[get_weather(city="Paris")]<|tool_call_end|>

Pass tools= to apply_chat_template and they are rendered into the system prompt.

Links

• Project website: mlx-optiq.com
• All OptiQ quants: mlx-optiq.com/models
• Base model: LiquidAI/LFM2.5-1.2B-Instruct