Skip to content
localmodel.run

Text model · LFM

LF LFM2 24B-A2B: RAM and VRAM requirements

LFM2 24B-A2B needs about 15.4 GB to run at Q4_K_M (the Q4_K_M GGUF file is ~13.43 GB to download; KV cache and overhead add the rest), or about 25.6 GB at Q8_0. The lightest hardware that runs it is Apple M4 (24GB).

LFM family · 24B params (Mixture-of-Experts: activates only 2.3B of 24B params per token, so generation is faster than the total size suggests) · released Feb 2026.

Shopping for hardware? See what runs LFM2 24B-A2B →

License LFM Open License v1.0 · Conditional ↓ 3.7K/mo ♥ 340 on HuggingFace
Q4_K_M GGUF
13.43 GB
Q8_0 GGUF
23.61 GB
Memory @ Q4 (4k)
~15.4 GB
Context
32 k

Q4_K_M and Q8_0 are GGUF download (file) sizes. Memory @ Q4 includes the KV cache and overhead, so it runs larger than the file on disk.

Will it run on your device?

LFM2 24B-A2B runs on 15 of 40 tracked devices at Q4_K_M.

13 run well 2 tight fit 25 too small
Fit check Q4_K_M
No, not enough memory
needs 15.4 GB usable 10.5 GB
See the full breakdown

Memory breakdown

Weights (Q4_K_M)13.43 GB
+
KV cache (4k)1.2 GB
+
Overhead0.8 GB
=
Total15.4 GB

How context length changes it

4k context ~15.4 GB 32k context ~23.6 GB 128k context ~51.8 GB

Longer context grows the KV cache, which for LFM2 24B-A2B is sized by its 2.3B active params, not the full 24B. It needs ~15.4 GB at 4k and ~51.8 GB at 128k.

Quantization sizes

GGUF quantson disk
Quantization Size on disk
Q2_K 10.1 GB est
Q3_K_M 11.7 GB est
Q4_K_M (default) 13.43 GB
Q5_K_M 17.1 GB est
Q6_K 19.7 GB est
Q8_0 23.61 GB
FP16 44.42 GB

Lower quant = smaller and faster, slightly lower quality. Q4_K_M is the common default.

Run it

Ollama
$ ollama run lfm2:24b
llama.cpp
$ llama-cli -hf LiquidAI/LFM2-24B-A2B-GGUF:Q4_K_M
LM Studio
$ lms get LiquidAI/LFM2-24B-A2B-GGUF

Which devices can run LFM2 24B-A2B?

Apple Silicon Macs

RAM-only laptops

iPhone & iPad

Android

NVIDIA GPUs

AMD GPUs

Same job, different size

Pick by what fits your memory: step down to free up VRAM, or step up if you have headroom.

Next size down ERNIE 4.5 21B-A3B 21B Next size up Gemma 4 26B-A4B 26.5B

Similar models

Head-to-head

FAQ

How much VRAM or RAM does LFM2 24B-A2B need?

At Q4_K_M, LFM2 24B-A2B needs about 15.4 GB (weights ~13.43 GB + KV cache + overhead) at a 4k context. At Q8_0 budget ~25.6 GB.

What is the Q4_K_M GGUF file size of LFM2 24B-A2B?

The Q4_K_M GGUF file is about 13.43 GB to download, and the Q8_0 GGUF is about 23.61 GB. That is the weights file on disk; to run it you also need room for the KV cache and overhead, so budget ~15.4 GB of memory at Q4_K_M.

Can LFM2 24B-A2B run on a laptop?

LFM2 24B-A2B is large; you need a 24 GB+ GPU or a 32-48 GB Mac at Q4_K_M.

Is LFM2 24B-A2B cheaper to run because it is a MoE model?

It is faster, not lighter. LFM2 24B-A2B activates only 2.3B of 24B params per token (so it runs quickly), but all experts must stay in memory, so it still needs memory for the full 24B.

Can I use LFM2 24B-A2B commercially?

Conditionally. LFM Open License: free under $10M revenue; larger companies need a commercial license.

Understand the numbers

Short guides to the ideas behind LFM2 24B-A2B's memory and quant figures.

The largest LFM2 model: 24B total / 2.3B active MoE per Liquid's own spec table (the -A2B suffix rounds down). Fits ~32GB unified/system RAM at Q4 per their own benchmark. 32K native context, smaller than the newer LFM2.5 line already in this catalog. LFM Open License v1.0.

Sources

Last validated 2026-08-03. Memory figures are estimates. See methodology.