LFM2.5 230M needs about 1.1 GB of memory at Q4_K_M (a 0.2 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 8 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.
LFM2.5‑230M: compact hybrid model for on‑device text tasks
LFM2.5‑230M is a 230 M‑parameter hybrid model with 14 layers and a 32,768‑token context. It is optimized for edge inference, delivering fast throughput on low‑cost CPUs and GPUs. The model supports function calling and is suited for data extraction and lightweight agentic pipelines.
Good at
- Fast edge inference
- Function calling support
- Compact 230 M parameter size
Limits
- Not suited for advanced math or creative writing
- No vision capabilities
- Requires 32 k context for full performance
Best for
- On‑device data extraction
- Lightweight agentic pipelines
- Tool‑based workflows
Who it’s for. Developers building on‑device LLM applications that need quick inference and function‑calling. Ideal for edge devices with limited memory and compute resources.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_0 | 0.1 GB | 1.1 GB | Fits | Get file → | |
| Q4_0 | 0.1 GB | 1.1 GB | Fits | Get file → | |
| Q4_K_M | 0.2 GB | 1.1 GB | Fits | Get file → | |
| Q5_K_M | 0.2 GB | 1.1 GB | Fits | Get file → | |
| Q6_K | 0.2 GB | 1.2 GB | Fits | Get file → | |
| Q8_0 | 0.2 GB | 1.2 GB | Fits | Get file → | |
| BF16 | 0.5 GB | 1.4 GB | Fits | Get file → | |
| F16 | 0.5 GB | 1.4 GB | Fits | Get file → |
For 8 GB VRAM, use Q4_0 or Q4_K_M (≈0.15 GB file, ~1.1 GB memory). If 12 GB VRAM is available, Q5_K_M or Q6_K provide higher precision with ~0.17–0.19 GB file and ~1.1–1.2 GB memory. BF16 and F16 offer the best quality but need ~0.46 GB file and ~1.4 GB memory.
What GPU do you need for LFM2.5 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 BF16 (0.5 GB file, about 1.4 GB in use).
Which LFM2.5 230M quant should you choose?
Q4_K_M is the usual starting point for LFM2.5 230M: it balances file size and quality well. Here it is a 0.2 GB file.
The Q8_0 file is 0.2 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 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 230M
How much VRAM does LFM2.5 230M need?
About 1.1 GB at Q4_K_M 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 230M on an 8 GB GPU?
Yes, with BF16: 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 230M GGUF?
It depends on the quant: Q4_0 is 0.1 GB, Q4_0 is 0.1 GB, Q4_K_M is 0.2 GB, Q5_K_M is 0.2 GB, Q6_K is 0.2 GB, Q8_0 is 0.2 GB and 2 more.
What context length does LFM2.5 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 230M GGUF?
The files are on Hugging Face in the LiquidAI/LFM2.5-230M-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 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 LFM2.5‑230M on a Raspberry Pi 5 with 8 GB VRAM?
Yes, the Q4_0 or Q4_K_M quantizations fit within 8 GB VRAM, using about 1.1 GB of memory at 8 k context.
Does LFM2.5‑230M support image inputs?
No, the model is text‑only and does not include vision capabilities.
What is the maximum context length I can use?
The model supports up to 32,768 tokens in context.