Skip to content
localmodel.run

Text model · LFM

LF LFM3-1.7B: RAM and VRAM requirements

LFM3-1.7B needs about 2.1 GB to run at Q4_K_M (the Q4_K_M GGUF file is ~1.02 GB to download; KV cache and overhead add the rest), or about 2.8 GB at Q8_0. The lightest hardware that runs it is Nvidia GeForce RTX 2060 (6GB).

LFM family · 1.73B params · released Oct 2026.

Shopping for hardware? See what runs LFM3-1.7B →

Set up LFM3-1.7B on your hardware →

License LFM Open License v1.0 · Conditional ↓ 37.6K/mo ♥ 663 on HuggingFace
Q4_K_M GGUF
1.02 GB
Q8_0 GGUF
1.72 GB
Memory @ Q4 (4k)
~2.1 GB
Context
256 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?

LFM3-1.7B 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 2.1 GB usable 10.5 GB
See the full breakdown
$ollama run lfm3:1.7b

Memory breakdown

Weights (Q4_K_M)1.02 GB
+
KV cache (4k)0.3 GB
+
Overhead0.8 GB
=
Total2.1 GB

How context length changes it

4k context ~2.1 GB 32k context ~4.3 GB 128k context ~11.9 GB

LFM3-1.7B is small enough that the context window is the thing to size for: 128k needs ~11.9 GB versus ~2.1 GB at 4k, so pick the window you actually use.

Speed drops too: every token re-reads the KV cache, so at 128k context LFM3-1.7B generates at roughly ~9% of its short-context speed. How this is estimated.

Quantization sizes

GGUF quantson disk
Quantization Size on disk
Q2_K 0.7 GB est
Q3_K_M 0.8 GB est
Q4_K_M (default) 1.02 GB
Q5_K_M 1.2 GB est
Q6_K 1.4 GB est
Q8_0 1.72 GB
FP16 3.23 GB

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

Run it

Ollama
$ ollama run lfm3:1.7b
llama.cpp
$ llama-cli -hf LiquidAI/LFM3-1.7B-GGUF:Q4_K_M
LM Studio
$ lms get LiquidAI/LFM3-1.7B-GGUF

Which devices can run LFM3-1.7B?

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 Qwen3 1.7B 1.7B Next size up LFM3-VL-2B 1.92B

Similar models

Head-to-head

FAQ

How much VRAM or RAM does LFM3-1.7B need?

At Q4_K_M, LFM3-1.7B needs about 2.1 GB (weights ~1.02 GB + KV cache + overhead) at a 4k context. At Q8_0 budget ~2.8 GB.

What is the Q4_K_M GGUF file size of LFM3-1.7B?

The Q4_K_M GGUF file is about 1.02 GB to download, and the Q8_0 GGUF is about 1.72 GB. That is the weights file on disk; to run it you also need room for the KV cache and overhead, so budget ~2.1 GB of memory at Q4_K_M.

Can LFM3-1.7B run on a laptop?

Yes, LFM3-1.7B fits on a 16 GB machine at Q4_K_M and runs on Apple Silicon or a 12 GB+ GPU comfortably.

Can I use LFM3-1.7B 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 LFM3-1.7B's memory and quant figures.

Liquid hybrid convolution/attention text model with 1.73B parameters and a 262,144-token vendor context. Measured official GGUF files: Q4_K_M 1,099,627,808 bytes, Q8_0 1,846,238,496 bytes and BF16 3,466,626,336 bytes, converted to GiB. Ollama Q4 matches the vendor file. Context memory uses the existing conservative calculator estimate.

Sources

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