LFM2.5 1.2B Thinking heretic 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 15 quants, from 0.5 GB (Q2_K) to 2.3 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.
1.2B on‑device reasoning model with tool‑calling
LFM2.5 1.2B Thinking heretic is a 1.17B parameter hybrid conv+attention model. It supports reasoning, tool use, and multilingual text generation. The model is optimized for local inference with low memory footprints.
Good at
- Reasoning and chain‑of‑thought
- Tool calling and function execution
- Multilingual text generation
Limits
- No vision support
- Not suited for heavy knowledge‑intensive tasks
- Requires at least 2.1 GB VRAM for 8 k context
Best for
- Local agents needing reasoning
- Math and multi‑step problem solving
- RAG and structured data extraction
Who it’s for. Developers who need a reasoning model that fits on a phone or edge device. The model runs under 1 GB VRAM on an AMD Ryzen CPU and supports llama.cpp, MLX, and vLLM.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 0.5 GB | 2.1 GB | Fits | Get file → | |
| IQ3_S | 0.6 GB | 2.1 GB | Fits | Get file → | |
| Q3_K_S | 0.6 GB | 2.1 GB | Fits | Get file → | |
| Q3_K_M | 0.6 GB | 2.2 GB | Fits | Get file → | |
| Q3_K_L | 0.6 GB | 2.2 GB | Fits | Get file → | |
| IQ4_XS | 0.7 GB | 2.2 GB | Fits | Get file → | |
| Q4_0 | 0.7 GB | 2.3 GB | Fits | Get file → | |
| Q4_K_S | 0.7 GB | 2.3 GB | Fits | Get file → | |
| Q4_K_M | 0.7 GB | 2.3 GB | Fits | Get file → | |
| Q4_1 | 0.8 GB | 2.3 GB | Fits | Get file → | |
| Q5_K_S | 0.8 GB | 2.4 GB | Fits | Get file → | |
| Q5_K_M | 0.8 GB | 2.4 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 → | |
| F16 | 2.3 GB | 3.9 GB | Fits | Get file → |
For 8 GB VRAM, Q4_K_M (0.73 GB) or Q5_K_S (0.83 GB) provide a good balance of speed and memory. If memory is tighter, Q3_K_M (0.6 GB) or Q3_K_S (0.56 GB) are viable. The smallest quant, Q2_K (0.48 GB), is available for extreme memory constraints.
What GPU do you need for LFM2.5 1.2B Thinking heretic?
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 (2.3 GB file, about 3.9 GB in use).
Which LFM2.5 1.2B Thinking heretic quant should you choose?
Q4_K_M is the usual starting point for LFM2.5 1.2B Thinking heretic: 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.
If memory is tight, the smallest file is Q2_K at 0.5 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about LFM2.5 1.2B Thinking heretic
How much VRAM does LFM2.5 1.2B Thinking heretic 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 LFM2.5 1.2B Thinking heretic on an 8 GB GPU?
Yes, with F16: the 2.3 GB file needs about 3.9 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of LFM2.5 1.2B Thinking heretic GGUF?
It depends on the quant: Q2_K is 0.5 GB, IQ3_S is 0.6 GB, Q3_K_S is 0.6 GB, Q3_K_M is 0.6 GB, Q3_K_L is 0.6 GB, IQ4_XS is 0.7 GB and 9 more.
What context length does LFM2.5 1.2B Thinking heretic 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 LFM2.5 1.2B Thinking heretic GGUF?
The files are on Hugging Face in the saidutta69/LFM2.5-1.2B-Thinking-heretic 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 1.2B Thinking heretic 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 1.2B on 8 GB VRAM?
Yes. The Q4_K_M quant uses about 2.3 GB at 8 k context, well below 8 GB.
Does the model support vision tasks?
No, it is a text‑only model.
What inference engines are supported?
It runs on llama.cpp, MLX, and vLLM.