LFM2.5 VL 3B needs about 4.2 GB of memory at Q4_K_M (a 1.7 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 25 quants, from 0.0 GB to 5.4 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 8.1 GB.
LFM2.5 VL 3B – multimodal 3B‑parameter model for on‑device vision and text
LFM2.5 VL 3B is a 3B‑parameter multimodal model built on LFM2.5‑2.6B with a SigLIP2 NaFlex vision encoder. It supports text and image input, offers OCR with layout annotation, and runs efficiently on devices with under 3.3 GB of memory at 228 tok/s on an Apple M5 Max.
Good at
- Vision‑text grounding
- Full‑page OCR with layout
- Low‑latency inference on edge devices
Limits
- Not suited for long‑context reasoning
- No support for languages beyond listed set
- Requires mmproj file for vision
Best for
- Real‑time object detection in automotive
- Batch OCR of scanned documents
- On‑device translation of menus and signs
Who it’s for. Developers building edge applications that need both text and image understanding. Ideal for mobile or embedded systems where memory and latency are constrained.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 0.0 GB | 2.5 GB | Fits | Get file → | |
| IQ2_M | 1.0 GB | 3.5 GB | Fits | Get file → | |
| Q2_K | 1.1 GB | 3.6 GB | Fits | Get file → | |
| IQ3_XXS | 1.1 GB | 3.6 GB | Fits | Get file → | |
| Q2_K_L | 1.2 GB | 3.7 GB | Fits | Get file → | |
| IQ3_XS | 1.2 GB | 3.7 GB | Fits | Get file → | |
| Q3_K_S | 1.3 GB | 3.8 GB | Fits | Get file → | |
| IQ3_M | 1.3 GB | 3.8 GB | Fits | Get file → | |
| Q3_K_M | 1.4 GB | 3.9 GB | Fits | Get file → | |
| Q3_K_L | 1.5 GB | 4.0 GB | Fits | Get file → | |
| Q3_K_XL | 1.5 GB | 4.0 GB | Fits | Get file → | |
| IQ4_XS | 1.5 GB | 4.0 GB | Fits | Get file → | |
| IQ4_NL | 1.6 GB | 4.1 GB | Fits | Get file → | |
| Q4_0 | 1.6 GB | 4.1 GB | Fits | Get file → | |
| Q4_K_S | 1.6 GB | 4.1 GB | Fits | Get file → | |
| Q4_K_M | 1.7 GB | 4.2 GB | Fits | Get file → | |
| Q4_K_L | 1.7 GB | 4.3 GB | Fits | Get file → | |
| Q4_1 | 1.7 GB | 4.3 GB | Fits | Get file → | |
| Q5_K_S | 1.9 GB | 4.4 GB | Fits | Get file → | |
| Q5_K_M | 1.9 GB | 4.5 GB | Fits | Get file → | |
| Q5_K_L | 2.0 GB | 4.5 GB | Fits | Get file → | |
| Q6_K | 2.2 GB | 4.8 GB | Fits | Get file → | |
| Q6_K_L | 2.3 GB | 4.8 GB | Fits | Get file → | |
| Q8_0 | 2.9 GB | 5.4 GB | Fits | Get file → | |
| BF16 | 5.4 GB | 7.9 GB | Tight | Get file → |
For 8 GB VRAM, Q4_K_M (1.68 GB) balances size and performance. With 4 GB VRAM, Q3_K_M (1.29 GB) or Q3_K_S (1.28 GB) are suitable. For 2 GB VRAM, use IQ2_M (1.02 GB) or Q2_K (1.1 GB).
What GPU do you need for LFM2.5 VL 3B?
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 IQ4_XS (1.5 GB file, about 4.0 GB in use).
- 6 GB VRAM: the best quant that fits is Q8_0 (2.9 GB file, about 5.4 GB in use).
- 8 GB VRAM: the best quant that fits is BF16 (5.4 GB file, about 7.9 GB in use).
Which LFM2.5 VL 3B quant should you choose?
Q4_K_M is the usual starting point for LFM2.5 VL 3B: it balances file size and quality well. Here it is a 1.7 GB file.
The Q8_0 file is 2.9 GB, about 1.7 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 the first one in the table at 0.0 GB, with the quality trade-offs that come with fewer bits.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about LFM2.5 VL 3B
How much VRAM does LFM2.5 VL 3B need?
About 4.2 GB at Q4_K_M with an 8K context window. That is the 1.7 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 LFM2.5 VL 3B on an 8 GB GPU?
Yes, with BF16: the 5.4 GB file needs about 7.9 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of LFM2.5 VL 3B GGUF?
It depends on the quant: file is 0.0 GB, IQ2_M is 1.0 GB, Q2_K is 1.1 GB, IQ3_XXS is 1.1 GB, Q2_K_L is 1.2 GB, IQ3_XS is 1.2 GB and 19 more.
What context length does LFM2.5 VL 3B support?
The GGUF header lists a trained context length of 128,000 tokens (125K). The KV cache grows with the context you use: about 2.0 GB at 8K and 8.1 GB at 32K.
Where can I download LFM2.5 VL 3B GGUF?
The files are on Hugging Face in the bartowski/LiquidAI_LFM2.5-VL-3B-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 LFM2.5 VL 3B 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 LFM2.5 VL 3B on 8 GB VRAM?
Yes, the Q4_K_M quant (1.68 GB) fits comfortably in 8 GB VRAM and offers good performance.
Does the model support OCR of full pages?
Yes, it provides full‑page OCR with layout annotation as a built‑in capability.
What vision encoder does it use?
It uses the SigLIP2 NaFlex shape‑optimized 400M vision encoder.