LiquidAI_LFM2.5 1.2B Instruct needs about 2.3 GB of memory at Q4_K_M (a 0.7 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 3 quants, from 0.7 GB (Q4_K_M) to 1.2 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.
LiquidAI LFM2.5 1.2B Instruct – a compact, text‑only model for on‑device inference
LiquidAI LFM2.5 1.2B Instruct is a 1.2B parameter hybrid model built on the LFM2 architecture. It supports a 32,768 token context and is optimized for edge inference with llama.cpp, MLX, and vLLM. The model excels in agentic tasks, data extraction, and RAG.
Good at
- Agentic tasks
- Data extraction
- RAG
Limits
- No vision support
- Not suited for knowledge‑intensive or programming tasks
- Requires at least 2.3 GB VRAM for Q4_K_M at 8k context
Best for
- Chatbots on mobile
- Data extraction pipelines
- RAG applications
Who it’s for. Developers building lightweight, on‑device chat or data‑extraction applications that need a small footprint and fast inference. It is suitable for users who prefer text‑only models and do not require extensive programming or knowledge‑heavy use.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K_M | 0.7 GB | 2.3 GB | Fits | Get file → | |
| Q6_K | 1.0 GB | 2.5 GB | Fits | Get file → | |
| Q8_0 | 1.2 GB | 2.8 GB | Fits | Get file → |
For 8 GB VRAM, Q4_K_M (0.73 GB) is recommended, using about 2.3 GB of VRAM at 8k context. If more memory is available, Q6_K (0.96 GB) uses roughly 2.5 GB. Q8_0 (1.25 GB) requires about 2.8 GB and offers the highest precision.
What GPU do you need for LiquidAI_LFM2.5 1.2B Instruct?
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 (1.2 GB file, about 2.8 GB in use).
Which LiquidAI_LFM2.5 1.2B Instruct quant should you choose?
Q4_K_M is the usual starting point for LiquidAI_LFM2.5 1.2B Instruct: it balances file size and quality well. Here it is a 0.7 GB file.
The Q8_0 file is 1.2 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.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about LiquidAI_LFM2.5 1.2B Instruct
How much VRAM does LiquidAI_LFM2.5 1.2B Instruct need?
About 2.3 GB at Q4_K_M with an 8K context window. That is the 0.7 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 LiquidAI_LFM2.5 1.2B Instruct on an 8 GB GPU?
Yes, with Q8_0: the 1.2 GB file needs about 2.8 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of LiquidAI_LFM2.5 1.2B Instruct GGUF?
It depends on the quant: Q4_K_M is 0.7 GB, Q6_K is 1.0 GB, Q8_0 is 1.2 GB.
What context length does LiquidAI_LFM2.5 1.2B Instruct support?
The GGUF header lists a trained context length of 128,000 tokens (125K). The KV cache grows with the context you use: about 1.1 GB at 8K and 4.3 GB at 32K.
Where can I download LiquidAI_LFM2.5 1.2B Instruct GGUF?
The files are on Hugging Face in the lmstudio-community/LFM2.5-1.2B-Instruct-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 LiquidAI_LFM2.5 1.2B Instruct 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 LiquidAI LFM2.5 1.2B Instruct on 8 GB VRAM?
Yes, the Q4_K_M quantization uses about 2.3 GB of VRAM at 8k context, well within an 8 GB budget.
Does this model support image inputs?
No, it is a text‑only model with no vision capabilities.
What inference engines can I use?
It is supported by llama.cpp, MLX, and vLLM for edge inference.