Skip to content
localmodel.run

Text model · LFM

LF LFM3 Omni 2B: RAM and VRAM requirements

LFM3 Omni 2B needs about 3.1 GB to run at Q4_K_M (the Q4_K_M GGUF file is ~1.91 GB to download; KV cache and overhead add the rest), or about 3.8 GB at Q8_0. The lightest hardware that runs it is Nvidia GeForce RTX 2060 (6GB).

LFM family · 2.17B params · released Oct 2026.

Shopping for hardware? See what runs LFM3 Omni 2B →

Vision · accepts images License LFM Open License v1.0 · Conditional ↓ 8.4K/mo ♥ 258 on HuggingFace
Q4_K_M GGUF
1.91 GB
Q8_0 GGUF
2.6 GB
Memory @ Q4 (4k)
~3.1 GB
Context
125 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?

LFM3 Omni 2B runs on 49 of 49 tracked devices at Q4_K_M.

49 run well 0 tight fit 0 too small
Fit check Q4_K_M
Yes, it runs fast
needs 3.1 GB usable 10.5 GB
See the full breakdown
$ollama run lfm3-omni:2b

Memory breakdown

Weights (Q4_K_M)1.91 GB
+
KV cache (4k)0.4 GB
+
Overhead0.8 GB
=
Total3.1 GB

How context length changes it

4k context ~3.1 GB 32k context ~5.5 GB 128k context ~14 GB

LFM3 Omni 2B is small enough that the context window is the thing to size for: 128k needs ~14 GB versus ~3.1 GB at 4k, so pick the window you actually use.

Speed drops too: every token re-reads the KV cache, so at 128k context LFM3 Omni 2B generates at roughly ~14% of its short-context speed. How this is estimated.

Quantization sizes

GGUF quantson disk
Quantization Size on disk
Q2_K 0.9 GB est
Q3_K_M 1.1 GB est
Q4_K_M (default) 1.91 GB
Q5_K_M 1.5 GB est
Q6_K 1.8 GB est
Q8_0 2.6 GB
FP16 4.11 GB

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

Run it

Ollama
$ ollama run lfm3-omni:2b
llama.cpp
$ llama-cli -hf LiquidAI/LFM3-Omni-2B-GGUF:Q4_K_M
LM Studio
$ lms get LiquidAI/LFM3-Omni-2B-GGUF

Which devices can run LFM3 Omni 2B?

NVIDIA GPUs

Apple Silicon Macs

RAM-only laptops

iPhone & iPad

Android

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 Sarvam-1 2B 2B Next size up Granite 3.1 2B 2.53B

Similar models

Head-to-head

FAQ

How much VRAM or RAM does LFM3 Omni 2B need?

At Q4_K_M, LFM3 Omni 2B needs about 3.1 GB (weights ~1.91 GB + KV cache + overhead) at a 4k context. At Q8_0 budget ~3.8 GB.

What is the Q4_K_M GGUF file size of LFM3 Omni 2B?

The Q4_K_M GGUF file is about 1.91 GB to download, and the Q8_0 GGUF is about 2.6 GB. That is the weights file on disk; to run it you also need room for the KV cache and overhead, so budget ~3.1 GB of memory at Q4_K_M.

Can LFM3 Omni 2B run on a laptop?

Yes, LFM3 Omni 2B fits on a 16 GB machine at Q4_K_M and runs on Apple Silicon or a 12 GB+ GPU comfortably.

Can I use LFM3 Omni 2B commercially?

Conditionally. LFM Open License v1.0: free for organizations below $10M annual revenue; larger organizations require separate terms.

Understand the numbers

Short guides to the ideas behind LFM3 Omni 2B's memory and quant figures.

Liquid omni model, 2.17B total parameters. Measured official GGUF totals include the full BF16 vision/audio projector (949,841,248 bytes): Q4_K_M 2,049,471,328 bytes, Q8_0 2,796,082,016 bytes, BF16 4,416,469,856 bytes, converted to GiB. The vendor text config specifies 128,000 tokens; the prose card states 32K. This fit estimates text/image use with the complete encoder weights, not measured speech-pipeline peak memory. llama.cpp v0.8.0 or newer supports the model; the vendor requires reasoning off.

Sources

Catalog updated 2026-11-01. Memory figures are estimates. See methodology.