LFM2.5 Encoder 230M needs about 1.2 GB of memory at Q8_0 (a 0.2 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 3 quants, from 0.1 GB (Q4_0) to 0.5 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 1.9 GB.
LFM2.5-Encoder 230M – lightweight multilingual bidirectional encoder
LFM2.5-Encoder 230M is a 230 million parameter multilingual encoder built on the LFM2 architecture. It offers 8 k context and is designed for fine‑tuning into classification, token classification, retrieval, reranking, and semantic similarity tasks across 15 languages.
Good at
- Bidirectional masked language modeling
- Fine‑tuning for classification and retrieval
- Runs efficiently on-device
Limits
- No vision support
- License type is other
- No mixture of experts
Best for
- On‑device text classification
- Semantic similarity search
- Multilingual NLI tasks
Who it’s for. Developers needing a small, fast encoder for on‑device inference or fine‑tuning into downstream models. Suitable for edge deployments and CPU‑only environments.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_0 | 0.1 GB | 1.1 GB | Fits | Get file → | |
| Q8_0 | 0.2 GB | 1.2 GB | Fits | Get file → | |
| F16 | 0.5 GB | 1.4 GB | Fits | Get file → |
For 8 GB VRAM, Q4_0 (0.15 GB) uses about 1.1 GB memory at 8 k context. Q8_0 (0.25 GB) uses about 1.2 GB. If higher precision is needed, F16 (0.46 GB) uses about 1.4 GB. Choose the smallest quant that meets your precision needs.
What GPU do you need for LFM2.5 Encoder 230M?
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 F16 (0.5 GB file, about 1.4 GB in use).
Which LFM2.5 Encoder 230M quant should you choose?
Q8_0 (0.2 GB) is the middle option in this repository. Close to lossless. It costs roughly twice the memory of a 4-bit file, so it suits cases where you have plenty of VRAM.
If memory is tight, the smallest file is Q4_0 at 0.1 GB, with the quality trade-offs that come with fewer bits. The usual sweet spot between file size and quality, and the quant most people start with.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about LFM2.5 Encoder 230M
How much VRAM does LFM2.5 Encoder 230M need?
About 1.2 GB at Q8_0 with an 8K context window. That is the 0.2 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 Encoder 230M on an 8 GB GPU?
Yes, with F16: the 0.5 GB file needs about 1.4 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of LFM2.5 Encoder 230M GGUF?
It depends on the quant: Q4_0 is 0.1 GB, Q8_0 is 0.2 GB, F16 is 0.5 GB.
What context length does LFM2.5 Encoder 230M support?
The GGUF header lists a trained context length of 128,000 tokens (125K). The KV cache grows with the context you use: about 0.5 GB at 8K and 1.9 GB at 32K.
Where can I download LFM2.5 Encoder 230M GGUF?
The files are on Hugging Face in the LiquidAI/LFM2.5-Encoder-230M-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 Encoder 230M 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-Encoder 230M on 8 GB VRAM?
Yes. The Q4_0 quant uses about 1.1 GB memory at 8 k context, well within an 8 GB VRAM budget.
Does this model support vision tasks?
No. It is a text‑only encoder with no vision capabilities.
What languages can it handle?
It supports 15 languages: English, German, Spanish, French, Italian, Dutch, Polish, Portuguese, Arabic, Hindi, Japanese, Russian, Turkish, Vietnamese, Chinese.