LLMs · Liquid AI

LFM2.5 1.2B Instruct

Name: LFM2.5 1.2B Instruct
Author: Liquid AI

Parameters: 1.2B
Architecture: Dense
Max context: 125K tokens
License: lfm1.0
Released: 2026-01

How to run it

Quantized files · GGUF

Works in LM Studio, llama.cpp, Jan and Ollama. No sign-in needed.

huggingface.co/LiquidAI/LFM2.5-1.2B-Instruct-GGUF (opens in a new tab)

ollama run hf.co/LiquidAI/LFM2.5-1.2B-Instruct-GGUF

Original weights · Hugging Face

Full-precision files from Liquid AI, for fine-tuning, vLLM, Transformers.

huggingface.co/LiquidAI/LFM2.5-1.2B-Instruct (opens in a new tab)

No sign-in needed.

Official repo · GitHub

Reference code, inference scripts and docs from Liquid AI. The weights themselves download from Hugging Face.

github.com/Liquid4All/cookbook (opens in a new tab)

git clone https://github.com/Liquid4All/cookbook

Liquid AI’s official cookbook for running LFM models.

Links checked 6 days ago against Hugging Face, Ollama and GitHub.

Memory needed

Weights plus KV cache and runtime overhead. Leave a little extra free for the operating system and other apps.

KV cache: hybrid convolution + attention: only the attention layers keep a KV cache.

Quantization	Quality	Weights	Total at 8K context	Total at 32K context	Download
`Q2_K`	Low	0.5 GB	1.2 GB	1.5 GB	GGUF files (opens in a new tab)
`Q3_K_M`	Low	0.6 GB	1.3 GB	1.6 GB	GGUF files (opens in a new tab)
`Q4_K_M`	Good	0.7 GB	1.4 GB	1.7 GB	GGUF (opens in a new tab)
`Q5_K_M`	High	0.8 GB	1.5 GB	1.8 GB	GGUF (opens in a new tab)
`Q6_K`	High	1.0 GB	1.7 GB	2.0 GB	GGUF (opens in a new tab)
`Q8_0`	Very high	1.2 GB	1.9 GB	2.2 GB	GGUF (opens in a new tab)
`F16`	Full	2.3 GB	3.0 GB	3.3 GB	GGUF (opens in a new tab) Original weights (opens in a new tab)

Best hardware for LFM2.5 1.2B Instruct

The least expensive GPU in our catalog that runs it without offloading is the Intel Arc A380 (launch price $139) at Q8_0.

Grade · points	Hardware	Quant	Memory	Speed	Fit
64 ptsGood (B) · 64 of 100 points: Model quality 14.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 9.9/15	NVIDIA GeForce RTX 4060	`F16`	3.4 GB / 8.0 GB	67.0 tok/s	Runs well
61 ptsGood (B) · 61 of 100 points: Model quality 14.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.0/15	Intel Arc B580	`F16`	3.6 GB / 12.0 GB	85.7 tok/s	Runs well
61 ptsGood (B) · 61 of 100 points: Model quality 14.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.0/15	NVIDIA GeForce RTX 3060 12GB	`F16`	3.6 GB / 12.0 GB	88.6 tok/s	Runs well
61 ptsGood (B) · 61 of 100 points: Model quality 14.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.0/15	NVIDIA GeForce RTX 4070 SUPER	`F16`	3.6 GB / 12.0 GB	124 tok/s	Runs well
60 ptsGood (B) · 60 of 100 points: Model quality 14.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.2/15	Apple M4 Pro (24 GB)	`F16`	4.3 GB / 16.1 GB	57.9 tok/s	Runs well
60 ptsGood (B) · 60 of 100 points: Model quality 14.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.0/15	Apple M4 (16 GB)	`Q8_0`	2.8 GB / 10.7 GB	55.2 tok/s	Runs well
59 ptsGood (B) · 59 of 100 points: Model quality 14.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.5/15	AMD Radeon RX 9070 XT	`F16`	3.8 GB / 16.0 GB	136 tok/s	Runs well
59 ptsGood (B) · 59 of 100 points: Model quality 14.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.5/15	NVIDIA GeForce RTX 4060 Ti 16GB	`F16`	3.8 GB / 16.0 GB	70.9 tok/s	Runs well
59 ptsGood (B) · 59 of 100 points: Model quality 14.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.5/15	NVIDIA GeForce RTX 4070 Ti SUPER	`F16`	3.8 GB / 16.0 GB	165 tok/s	Runs well
59 ptsGood (B) · 59 of 100 points: Model quality 14.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.5/15	NVIDIA GeForce RTX 4080 SUPER	`F16`	3.8 GB / 16.0 GB	181 tok/s	Runs well

Check it on your own hardware

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.