LF

LFM2 2.6B Exp Heretic

LfmGGUF by TrisertUpdated 2025-12-31
Download on Hugging Face
Parameters2.57B
Context125K
Layers30
Architecturelfm2
Smallest quant0.1 GB
Quick answer

LFM2 2.6B Exp Heretic needs about 4.1 GB of memory at Q4_K_M (a 1.6 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 24 quants, from 0.1 GB (IQ2_M) to 5.1 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 8.1 GB.

LFM2 2.6B Exp Heretic – a 2.57B parameter LFM model with tool usage

LFM2 2.6B Exp Heretic is a 2.57B parameter language model based on the LFM2 architecture. It supports tool usage and was trained on 128000 context tokens. The model is available in multiple GGUF quantizations.

Good at

  • Tool usage
  • Large context handling
  • Compact quant options

Limits

  • No vision support
  • License not specified
  • No mixture of experts

Best for

  • Chatbots needing tool integration
  • Script generation with context limits
  • Low‑memory inference

Who it’s for. Developers building lightweight LLM applications that require tool interaction and can operate within limited GPU memory. The model is suitable for environments where a 2.57B parameter model is desired without large VRAM requirements.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ2_M0.1 GB 2.6 GB
Fits Get file →
IQ3_XXS0.2 GB 2.7 GB
Fits Get file →
TQ1_00.6 GB 3.1 GB
Fits Get file →
TQ2_00.7 GB 3.3 GB
Fits Get file →
Q2_K1.0 GB 3.5 GB
Fits Get file →
IQ3_XS1.1 GB 3.6 GB
Fits Get file →
IQ3_S1.2 GB 3.7 GB
Fits Get file →
Q3_K_S1.2 GB 3.7 GB
Fits Get file →
IQ3_M1.2 GB 3.7 GB
Fits Get file →
Q3_K_M1.3 GB 3.8 GB
Fits Get file →
Q3_K_L1.3 GB 3.9 GB
Fits Get file →
IQ4_XS1.4 GB 3.9 GB
Fits Get file →
Q4_01.5 GB 4.0 GB
Fits Get file →
IQ4_NL1.5 GB 4.0 GB
Fits Get file →
Q4_K_S1.5 GB 4.0 GB
Fits Get file →
Q4_K_M1.6 GB 4.1 GB
Fits Get file →
Q4_11.6 GB 4.1 GB
Fits Get file →
Q5_01.8 GB 4.3 GB
Fits Get file →
Q5_K_S1.8 GB 4.3 GB
Fits Get file →
Q5_K_M1.8 GB 4.3 GB
Fits Get file →
Q5_11.9 GB 4.5 GB
Fits Get file →
Q6_K2.1 GB 4.6 GB
Fits Get file →
Q8_02.7 GB 5.2 GB
Fits Get file →
F165.1 GB 7.7 GB
Tight Get file →

For 4‑GB VRAM, use IQ2_M (0.1 GB) or IQ3_XXS (0.17 GB). For 8‑GB VRAM, IQ4_K_M (1.56 GB) or Q5_K_M (1.83 GB) are suitable. For 12‑GB VRAM, Q6_K (2.11 GB) or Q8_0 (2.73 GB) provide higher quality. Memory estimates are approximate.

What GPU do you need for LFM2 2.6B Exp 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 Q4_K_S (1.5 GB file, about 4.0 GB in use).
  • 6 GB VRAM: the best quant that fits is Q8_0 (2.7 GB file, about 5.2 GB in use).
  • 8 GB VRAM: the best quant that fits is F16 (5.1 GB file, about 7.7 GB in use).

Which LFM2 2.6B Exp Heretic quant should you choose?

Q4_K_M is the usual starting point for LFM2 2.6B Exp Heretic: it balances file size and quality well. Here it is a 1.6 GB file.

The Q8_0 file is 2.7 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 IQ2_M at 0.1 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

Architecturelfm2
Parameters2.57B
Layers30
Trained context128,000 tokens
KV cache (fp16)240 KB per token

Source

Quantized byTrisert
FamilyLfm
Downloads (HF)770
Last updated2025-12-31

Frequently asked questions about LFM2 2.6B Exp Heretic

How much VRAM does LFM2 2.6B Exp Heretic need?

About 4.1 GB at Q4_K_M with an 8K context window. That is the 1.6 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 2.6B Exp Heretic on an 8 GB GPU?

Yes, with F16: the 5.1 GB file needs about 7.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of LFM2 2.6B Exp Heretic GGUF?

It depends on the quant: IQ2_M is 0.1 GB, IQ3_XXS is 0.2 GB, TQ1_0 is 0.6 GB, TQ2_0 is 0.7 GB, Q2_K is 1.0 GB, IQ3_XS is 1.1 GB and 18 more.

What context length does LFM2 2.6B Exp 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 2.0 GB at 8K and 8.1 GB at 32K.

Where can I download LFM2 2.6B Exp Heretic GGUF?

The files are on Hugging Face in the Trisert/LFM2-2.6B-Exp-heretic-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

Can I run LFM2 2.6B Exp Heretic on 8 GB VRAM?

Yes, use IQ4_K_M or Q5_K_M quantizations, which require about 4.1 GB or 4.3 GB of VRAM at 8k context.

Does this model support vision tasks?

No, the model has no vision capabilities.

What is the maximum context length for this model?

The model was trained on 128000 context tokens and supports that length.

Similar models

Browse all →

More from Trisert

All by Trisert →

More Lfm models

All Lfm →

Other tiny models

All tiny models →