LocalClaw/LLM/LFM3 VL 2B
Open-weight local LLM

LFM3 VL 2B

Use case: Reasoning. Check the memory fit and choose a setup path below.

8 GB short-context planning estimate8 GB RAMQ4_0Reasoning
Choose an app

This model needs a special runtime. Unsupported apps are clearly marked.

Compare models
LM StudioNot available for this model
UnslothNot available for this model
OllamaNot available for this model
Open on Hugging FaceFiles, licence and available downloads
llama.cppNot available for this model
Use with LocalClawOptional workspace after the model is installed
This model needs its official runtimellama.cpp containing 19cce6919992fdffd92e0ac152caaf1e2b8f254e (PR 31252); start with -c 4096; use the matching F16 mmproj, no video input
Parameters
1.92B
RAM planning estimate
8 GB
Model size
1.43 GB
Quantization
Q4_0

About LFM3 VL 2B

Catalogue summary: Liquid AI image understanding, OCR and visual reasoning model. Official QAD Q4_0 GGUF plus F16 vision projector totals 1.43GB. Plan for 8GB RAM at short context. Requires LFM3-enabled llama.cpp; video input is not yet supported.

LFM3 VL 2B has a 8 GB short-context RAM planning estimate based on its 1.43 GB GGUF. This is not a measured minimum or a guarantee at maximum context.

Use the official llama.cpp containing 19cce6919992fdffd92e0ac152caaf1e2b8f254e (PR 31252); start with -c 4096; use the matching F16 mmproj, no video input setup. The current low-bit files are not a stock LM Studio install.

Model sourcePublic GGUF repository (verified 2026-11-06): LiquidAI/LFM3-VL-2B-GGUFUpstream releaseLicence text
visionmultimodallightreasoning

Install path

01
Check RAM fitPlan for 8 GB RAM at short context; this is an estimate. Use Q4_0.
02
Load the modelFollow the official llama.cpp containing 19cce6919992fdffd92e0ac152caaf1e2b8f254e (PR 31252); start with -c 4096; use the matching F16 mmproj, no video input instructions. Stock LM Studio support is not confirmed.
03
Try your workloadStart with a short prompt, check memory use and adjust context length before running longer tasks.

Download formats for this model

These are quantizations of the same model. The RAM fit above uses Q4_0; larger files need more memory. File sizes exclude the runtime, context cache and optional vision projector.

Catalogue record

  • Family: lfm
  • Parameters: 1.92B
  • Recommended quantization: Q4_0
  • Short-context RAM estimate: 8 GB
  • Catalogue model size: 1.43 GB
  • Tags: vision, multimodal, light, reasoning

Practical limits

  • Catalogue RAM is a minimum estimate, not a guarantee for every context length or runtime.
  • Speed and memory use vary by quantization, backend, context length and system headroom.
  • Verify architecture, licence and usage restrictions in the linked upstream material before deployment.

Reading the scores

LocalClaw /10 is an editorial catalogue rating. Community /5 comes from member reviews. Your machine fit is a separate memory and runtime assessment.

Capability profile

Repository catalogue ratings used by LocalClaw's editorial rubric. They are not a standardized third-party benchmark.

speed
?
quality
?
coding
?
reasoning
?

Technical notes

Developer
Liquid AI
License
LFM 1.0
Context window
131,072 tokens
Architecture
24-layer tri-hybrid LFM3 backbone: 13 gated short convolutions, 5 sliding-window attention and 6 global attention layers. 1.73B language backbone, 164M vision encoder and 25M projector.

This model fits these next steps

Hardware fit is based on LocalClaw's RAM tier, model size and quantization metadata. Always leave memory headroom for your OS and runtime.

Entry laptop fitMacBook Air 8GBMore headroomMac mini M4 16GBAll compatible picks8GB RAM guide

Related catalogue entries

Linked mechanically by family, shared tags and nearby RAM tier; this is not a quality ranking.

LFM2.5-2.6B 2.7BLFM2.5-VL-3B 3B multimodalLFM2.5-8B-A1B 8.3B (1.5B active)LFM3 1.7B 1.73BLFM2.5-24B-A2B 24B (2.45B active, MoE)Agents-A1 4B 4BGemma 4 E4B E4BMiniCPM-V 4.7 (1B) 1B

Where to go next