LFM3-VL-2B needs about 3.2 GB of memory at Q4_K_M (a 1.1 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 6 quants, from 0.9 GB (IQ3_S) to 1.8 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 6.4 GB.
LFM3-VL-2B – a 1.73B vision‑language model for on‑device visual understanding
LFM3-VL-2B combines a 1.73B language model with a 164M vision encoder and a 25M projector. It accepts text, images, and video at native resolution and reasons step by step. The model supports 128K token context and 128,000 token vocabulary.
Good at
- Video question answering
- Spatial and embodied reasoning
- Step‑by‑step reasoning for complex visual tasks
Limits
- No audio input support
- Requires at least 3 GB VRAM for IQ4_XS at 8K context
- License is other, not open source
Best for
- On‑device visual QA
- Robotics perception tasks
- Educational video analysis
Who it’s for. Developers building offline vision‑language applications who need step‑by‑step reasoning and spatial understanding. The model is suitable for mobile or edge devices that can handle the required VRAM.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ3_S | 0.9 GB | 3.0 GB | Fits | Get file → | |
| IQ4_XS | 1.1 GB | 3.2 GB | Fits | Get file → | |
| Q4_K_M | 1.1 GB | 3.2 GB | Fits | Get file → | |
| Q5_K_M | 1.3 GB | 3.4 GB | Fits | Get file → | |
| Q6_K | 1.6 GB | 3.7 GB | Fits | Get file → | |
| Q8_0 | 1.8 GB | 4.0 GB | Fits | Get file → |
For 8 GB VRAM, use IQ4_XS (1.09 GB file, ~3.2 GB memory at 8K context) or Q4_K_M (1.13 GB file, ~3.2 GB memory). For tighter budgets, IQ3_S (0.92 GB file, ~3 GB memory) is a research‑grade option. Larger quants like Q6_K and Q8_0 provide higher fidelity at the cost of 3.7 GB and 4 GB memory respectively.
What GPU do you need for LFM3-VL-2B?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: the best quant that fits is Q8_0 (1.8 GB file, about 4.0 GB in use).
Which LFM3-VL-2B quant should you choose?
Q4_K_M is the usual starting point for LFM3-VL-2B: it balances file size and quality well. Here it is a 1.1 GB file.
The Q8_0 file is 1.8 GB, about 1.6 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
If memory is tight, the smallest file is IQ3_S at 0.9 GB, with the quality trade-offs that come with fewer bits. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about LFM3-VL-2B
How much VRAM does LFM3-VL-2B need?
About 3.2 GB at Q4_K_M with an 8K context window. That is the 1.1 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run LFM3-VL-2B on an 8 GB GPU?
Yes, with Q8_0: the 1.8 GB file needs about 4.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of LFM3-VL-2B GGUF?
It depends on the quant: IQ3_S is 0.9 GB, IQ4_XS is 1.1 GB, Q4_K_M is 1.1 GB, Q5_K_M is 1.3 GB, Q6_K is 1.6 GB, Q8_0 is 1.8 GB.
What context length does LFM3-VL-2B support?
The GGUF header lists a trained context length of 128,000 tokens (125K). The KV cache grows with the context you use: about 1.6 GB at 8K and 6.4 GB at 32K.
Where can I download LFM3-VL-2B GGUF?
The files are on Hugging Face in the AtomicChat/LFM3-VL-2B-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does LFM3-VL-2B use?
Hugging Face lists the license as other. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run LFM3-VL-2B on a device with 8 GB VRAM?
Yes, the IQ4_XS and Q4_K_M quants use about 3.2 GB of VRAM at 8K context, which fits within an 8 GB budget.
Does the model support audio input?
No, LFM3-VL-2B does not accept audio; it is a vision‑language model.
What languages can the model understand?
The model supports English, Arabic, Chinese, French, German, Italian, Japanese, Korean, Portuguese, Spanish, Vietnamese, Thai, Indonesian, Hindi, Russian, and Polish.