selfhost.directory

The index of self-hostable projects — ranked, categorized and served with live updates.
Own your data and stop giving it to big tech.

LiquidAI logo

LFM3-VL-2B

LiquidAI·Instruct·released 19 Oct 2026

lfm1.01.9B125K ctx ≈2.0 GB
Your GPU
ollama run lfm3-vl:2b
All 4 runners ↓

LiquidAI's LFM3-VL-2B is an instruct vision-language model with 1.9B parameters, built on the LiquidAI/LFM3-1.7B-Base model. It accepts text and images as input and produces text as output, with a context length of 128,000 tokens. It is released under the lfm1.0 licence, which allows limited commercial use. The page below covers its VRAM needs and how to run it locally.

How much VRAM does LFM3-VL-2B need?

At Q4_K_M LFM3-VL-2B needs about 2.0 GB with an 8K context (2.0 GB at Q4, 4.6 GB at full precision). Pick your GPU to see what fits and how fast it runs.

--vram
any
--gpus
--quant

Needs: ≈1.9 GB · Q4_K_M · 8K context — Pick a GPU to check the fit.

Change quant, context or KV cache to compare
Advanced: context 8K · KV cache 16-bit
Context
KV cache

Smallest GPU for each quant and context (KV cache 16-bit)

Quantisation4K8K16K32K64K
Q3_K_M
Q4_K_M
Q5_K_M
Q6_K
Q8_0
FP16

≤16 GB card24–32 GB48–96 GBmore

QuantisationQualityWeightsKV cacheTotal
Q2_KHeavy lossSmallest, noticeably worse qualityHeavy loss0.7 GB file0.4 GB≈1.6 GB
Q3_K_MNoticeable lossSmall, some quality lossNoticeable loss0.8 GB file0.4 GB≈1.7 GB
Q4_K_MrecommendedSmall lossRecommended default for local useSmall loss1.0 GB file0.4 GB≈1.9 GB
Q5_K_MNear-losslessNear-lossless for most tasksNear-lossless1.2 GB file0.4 GB≈2.1 GB
Q6_KNear-losslessPractically losslessNear-lossless1.3 GB file0.4 GB≈2.3 GB
Q8_0LosslessLossless in practice, twice the size of Q4Lossless1.7 GB file0.4 GB≈2.7 GB
FP16 / BF16LosslessFull precision, as releasedLossless3.2 GB file0.4 GB≈4.3 GB

Estimates for 8K tokens of context: weights + KV cache + runtime overhead. Speed assumes memory-bound decoding at your GPU’s bandwidth. Pick a GPU above and the fit column checks every row against it. Weights marked “file” use the real GGUF download size; “est.” rows are computed from the parameter count.

How to pick a quant

Take the biggest quant that fits with room left for context. Q4_K_M is the usual sweet spot. Below Q4 quality drops fast, above Q6 you mostly pay memory for nothing. Big models (70B+, large MoE) tolerate low quants better than small ones.

  • LosslessF16, Q8_0
  • Near-losslessQ6_K, Q5_K_M
  • Small lossQ5_K_S, Q4_K_M, IQ4_NL, IQ4_XS, Q4_K_S
  • Noticeable lossQ4_0, Q3_K_L, Q3_K_M, IQ3_M
  • Heavy lossIQ3_XXS, Q3_K_S, IQ2_M, Q2_K

Typical quality loss, not measured per model. Tiers come from KL divergence against FP16 on Llama 3 8B: llama.cpp perplexity README: LLaMA 3 8B scoreboard (KLD and ΔPPL vs FP16, Wikitext-2); llama.cpp quantize tool: ΔPPL per quant type @ Llama-3-8B

How much context fits?

Model quant
KV cache

At 8K context, Q4_K_M needs ≈1.9 GB(1.0 GB weights + 0.4 GB KV cache + 0.6 GB overhead). Pick a GPU above to see how much context fits.

024681K2K4K8K16K32K64K125KGB

Weights + overheadKV cache

Can the RTX 4060 run LFM3-VL-2B? →

Speed vs quality per quant

Smaller quants run faster and fit in less memory, but lose quality.

Q2_KQ3_K_MQ4_K_MQ5_K_MQ6_KQ8_0F16

Fits in VRAMTight fitCPU offload✕ won't runQuality is typical KL divergence vs FP16, not measured for this model.

Real-world speed of LFM3-VL-2B

Tokens per second measured by people running it at home. A GPU shows a median once 3 reports agree on the same setup. Reports far from our estimate wait for a review. Measure yours with llama-bench →

No medians yet. Be one of the first 3 to report a setup.

All reports

Run LFM3-VL-2B locally

--quantQ4_K_Mdefault; pick your GPU to tailor itLiquidAI/LFM3-VL-2B-GGUF

From the Ollama librarydocs
ollama run lfm3-vl:2b

The library build picks its own default quantisation.

Q4_K_M straight from Hugging Facedocs
ollama run hf.co/LiquidAI/LFM3-VL-2B-GGUF:Q4_K_M

Install Ollama

OpenAI-compatible server with a web UIdocs
llama-server -hf LiquidAI/LFM3-VL-2B-GGUF:Q4_K_M

Downloads the GGUF on first run. Installed through llama.app? The same flags work with `llama serve`.

Install llama.cpp

Download with the lms CLIdocs
lms get LiquidAI/LFM3-VL-2B-GGUF@Q4_K_M

Then load it from the LM Studio app or with `lms load`.

Install LM Studio

Ask about an imagedocs
python
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="LiquidAI/LFM3-VL-2B", dtype="auto", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/bee.jpg"},
            {"type": "text", "text": "Describe this image."},
        ],
    }
]
out = pipe(text=messages, max_new_tokens=256)
print(out[0]["generated_text"][-1]["content"])

Install Transformers

Quantised versions and downloads

RepositoryFormatQuantSizeDownloads
LiquidAI/LFM3-VL-2B-NVFP4SafetensorsNVFP4—1.4K
LiquidAI/LFM3-VL-2B-FP8SafetensorsFP8—1.2K
bartowski/LFM3-VL-2B-GGUFGGUFIQ2_M0.7 GB3.9K
bartowski/LFM3-VL-2B-GGUFGGUFQ2_K0.7 GB3.9K
bartowski/LFM3-VL-2B-GGUFGGUFIQ3_XXS0.8 GB3.9K
bartowski/LFM3-VL-2B-GGUFGGUFQ3_K_S0.8 GB3.9K
bartowski/LFM3-VL-2B-GGUFGGUFIQ3_XS0.8 GB3.9K
bartowski/LFM3-VL-2B-GGUFGGUFQ3_K_M0.8 GB3.9K
bartowski/LFM3-VL-2B-GGUFGGUFQ3_K_L0.9 GB3.9K
bartowski/LFM3-VL-2B-GGUFGGUFIQ3_M0.9 GB3.9K

About LFM3-VL-2B

LFM3-VL-2B is a vision-language model for on-device visual understanding. It takes text, images, and video as input, and combines the LFM3 language model with a new in-house vision encoder. New in LFM3-VL-2B: • Video understanding: the first LFM vision model that takes video as input. Ask questions about a recorded lecture, a walkthrough filmed on your phone, or a clip from a camera feed. • Thinking: the model reasons step by step before it answers. This helps most with harder visual questions, such as math, charts, and multi-step problems. • Spatial and embodied reasoning: the model finds objects and reasons about how they are placed relative to each other, a strong foundation for robotics applications. Find more information about the LFM3 family in our release blog post. The LFM3 family includes base, text, vision, audio, omni, and decision models.

Full model card on Hugging Face →

Alternatives to LFM3-VL-2B

All alternatives →
Qwen logo

Qwen3.5-2B

Qwen·Vision

≈2.2 GB

Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance.

2.3B256K ctx Apache 2.0
5.5MFeb 2026ollama
openbmb logo

MiniCPM-V-4.7-1B

openbmb·Vision

≈1.8 GB

MiniCPM-V 4.7 is the latest model in the MiniCPM-V series, available in a 1B variant. It builds on the family's single-image, multi-image, and video understanding capabilities, combining stronger foundation capabilities with efficient visual encoding and on-device inference.

1.6B256K ctx Apache 2.0
470.5KOct 2026gguf
Qwen logo≈2.6 GB

Meet Qwen3-VL — the most powerful vision-language model in the Qwen series to date.

2.1B256K ctx Apache 2.0
1.1MOct 2025ollama
LiquidAI logo

d1-2B

LiquidAI·Vision

≈2.0 GB

d1-2B is a decision model for text, pictures, and video, built on LFM3-VL-2B. It returns calibrated answers in one forward pass, without generating tokens.

1.9B125K ctx lfm1.0
16.2KOct 2026ollama
Qwen logo

Qwen3.5-2B-Base

Qwen·Vision

≈2.2 GB

Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance.

2.3B256K ctx Apache 2.0
322.9KFeb 2026gguf
meta-models logo

Muse-Glimmer-30B-assistant

meta-models·Vision

≈2.2 GB

Muse Glimmer is a 30-billion-parameter causal language model with a dedicated perception encoder, distilled from Muse Spark and purpose-built for autonomous agentic tasks on consumer hardware.

2.6B128K ctx Apache 2.0
129.6KAug 2026

LFM3-VL-2B FAQ

How much VRAM does LFM3-VL-2B need?
About 2.0 GB at Q4_K_M with an 8K context. Q8_0 needs about 2.9 GB and full precision about 4.6 GB. Longer contexts need more memory for the KV cache.
Can LFM3-VL-2B run on the RTX 4060?
Yes. It fits at FP16 / BF16 (about 4.3 GB), at an estimated 39 tokens/s. It fits comfortably on the Apple M4 (16 GB), Radeon RX 6700 XT, RTX 3060 12GB.
Can I use LFM3-VL-2B commercially?
Yes, with conditions. It is released under lfm1.0. Custom licence: read the model card before using it commercially.
What is the context length of LFM3-VL-2B?
128,000 tokens (125K). The full window needs far more memory than an 8K context.
How do I run LFM3-VL-2B with Ollama or llama.cpp?
With Ollama: ollama run lfm3-vl:2b. With llama.cpp: llama-server -hf LiquidAI/LFM3-VL-2B-GGUF:Q4_K_M, which starts an OpenAI-compatible server.

Data from Hugging Face, the model card and public leaderboards. Memory and speed figures are estimates; real numbers depend on the runtime, context length and drivers. Always check the licence yourself before commercial use.