LF

LFM2 24B A2B Heretic Ayun

LfmGGUF by mradermacherUpdated 2026-03-20
Download on Hugging Face
Parameters23.8B
Context125K
Layers40
Architecturelfm2moe
Smallest quant0.1 GB
Quick answer

LFM2 24B A2B Heretic Ayun needs about 17.6 GB of memory at Q4_K_M (a 14.4 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 24 quants, from 0.1 GB to 19.6 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 10.7 GB.

LFM2 24B A2B Heretic Ayun – a 23.8B parameter LFM2MOE model

LFM2 24B A2B Heretic Ayun is a 23.8B parameter LFM2MOE model with 2B active mixture‑of‑expert parameters. It supports a trained context of 128000 tokens and is optimized for tool usage. The model is available in multiple GGUF quantizations.

Good at

  • Tool usage
  • Large context handling
  • Mixture of experts

Limits

  • No vision support
  • License not specified
  • Large memory requirement

Best for

  • Tool‑centric applications
  • Long‑form reasoning
  • Complex dialogue

Who it’s for. Developers building AI assistants that need extensive context and tool integration. Researchers exploring mixture‑of‑expert architectures may also find it useful.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—0.1 GB 3.2 GB
Fits Get file →
IQ1_S4.8 GB 8.0 GB
Too big Get file →
IQ1_M5.4 GB 8.6 GB
Too big Get file →
IQ2_XXS6.3 GB 9.5 GB
Too big Get file →
IQ2_XS7.0 GB 10.2 GB
Too big Get file →
IQ2_S7.1 GB 10.3 GB
Too big Get file →
IQ2_M7.8 GB 11.0 GB
Too big Get file →
Q2_K_S8.1 GB 11.2 GB
Too big Get file →
Q2_K8.7 GB 11.9 GB
Too big Get file →
IQ3_XXS9.2 GB 12.4 GB
Too big Get file →
IQ3_XS9.8 GB 12.9 GB
Too big Get file →
IQ3_S10.3 GB 13.5 GB
Too big Get file →
Q3_K_S10.3 GB 13.5 GB
Too big Get file →
IQ3_M10.4 GB 13.6 GB
Too big Get file →
Q3_K_M11.4 GB 14.5 GB
Too big Get file →
Q3_K_L12.3 GB 15.5 GB
Too big Get file →
IQ4_XS12.7 GB 15.9 GB
Too big Get file →
Q4_013.5 GB 16.7 GB
Too big Get file →
Q4_K_S13.5 GB 16.7 GB
Too big Get file →
Q4_K_M14.4 GB 17.6 GB
Too big Get file →
Q4_114.9 GB 18.1 GB
Too big Get file →
Q5_K_S16.4 GB 19.6 GB
Too big Get file →
Q5_K_M16.9 GB 20.1 GB
Too big Get file →
Q6_K19.6 GB 22.8 GB
Too big Get file →

For 8 GB VRAM, use IQ1_S (file size 4.84 GB, memory 8 GB). If 10 GB VRAM is available, IQ1_M (5.38 GB, 8.6 GB) or IQ2_XXS (6.28 GB, 9.5 GB) are suitable. For 16 GB VRAM, Q4_K_M (14.42 GB, 17.6 GB) or Q5_K_M (16.92 GB, 20.1 GB) can be considered.

What GPU do you need for LFM2 24B A2B Heretic Ayun?

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 the main file (0.1 GB file, about 3.2 GB in use).
  • 6 GB VRAM: the best quant that fits is the main file (0.1 GB file, about 3.2 GB in use).
  • 8 GB VRAM: the best quant that fits is IQ1_S (4.8 GB file, about 8.0 GB in use).
  • 12 GB VRAM: the best quant that fits is Q2_K (8.7 GB file, about 11.9 GB in use).
  • 16 GB VRAM: the best quant that fits is IQ4_XS (12.7 GB file, about 15.9 GB in use).
  • 24 GB VRAM: the best quant that fits is Q6_K (19.6 GB file, about 22.8 GB in use).

Which LFM2 24B A2B Heretic Ayun quant should you choose?

Q4_K_M is the usual starting point for LFM2 24B A2B Heretic Ayun: it balances file size and quality well. Here it is a 14.4 GB file.

If memory is tight, the smallest file is the first one in the table at 0.1 GB, with the quality trade-offs that come with fewer bits.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturelfm2moe
Parameters23.8B (approximate, mixture of experts)
Active parameters2B
Layers40
Trained context128,000 tokens
KV cache (fp16)320 KB per token

Source

Quantized bymradermacher
FamilyLfm
Downloads (HF)601
Last updated2026-03-20

Frequently asked questions about LFM2 24B A2B Heretic Ayun

How much VRAM does LFM2 24B A2B Heretic Ayun need?

About 17.6 GB at Q4_K_M with an 8K context window. That is the 14.4 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 24B A2B Heretic Ayun on an 8 GB GPU?

Yes, with IQ1_S: the 4.8 GB file needs about 8.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of LFM2 24B A2B Heretic Ayun GGUF?

It depends on the quant: file is 0.1 GB, IQ1_S is 4.8 GB, IQ1_M is 5.4 GB, IQ2_XXS is 6.3 GB, IQ2_XS is 7.0 GB, IQ2_S is 7.1 GB and 18 more.

What context length does LFM2 24B A2B Heretic Ayun support?

The GGUF header lists a trained context length of 128,000 tokens (125K). The KV cache grows with the context you use: about 2.7 GB at 8K and 10.7 GB at 32K.

Where can I download LFM2 24B A2B Heretic Ayun GGUF?

The files are on Hugging Face in the mradermacher/LFM2-24B-A2B-heretic-ayun-i1-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 this model on 8 GB VRAM?

Yes, the IQ1_S quantization requires 8 GB VRAM and has an estimated memory usage of 8 GB at 8k context.

Does this model support vision tasks?

No, the model does not include vision capabilities.

What is the context length of this model?

The trained context length is 128000 tokens.

Other GGUFs of LFM2-24B-A2B-heretic-ayun

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Lfm models

All Lfm →

Other medium models

All medium models →