---
title: summerAIO/diffusion-lfm2.5-24b-a2b-blocktune
url: https://recentmodel.com/models/summerAIO/diffusion-lfm2.5-24b-a2b-blocktune/
status: archived · low-signal
published: 2026-11-11T19:52:08Z
updated: 2026-11-14T00:35:18Z
source: https://huggingface.co/summerAIO/diffusion-lfm2.5-24b-a2b-blocktune
---

# summerAIO/diffusion-lfm2.5-24b-a2b-blocktune

DLFM · BlockTune is an experimental block-diffusion derivative of summerAIO/diffusion-lfm2.5-24b-a2b. Mixture-of-experts, 24B total (64 experts, 4 per token). Context: 128K tokens (config). License: lfm-open-license-v1.0. Vendor-reported: "The measured BF16 merge had top-1 agreement of 82.81–96.88% across four probes, below the fixed 98% per-probe acceptance threshold."

- Vendor-reported: The measured BF16 merge had top-1 agreement of 82.81–96.88% across four probes, below the fixed 98% per-probe acceptance threshold.

- **Org:** summerAIO
- **Released:** 2026-11-11
- **Status:** archived · low-signal
- **License:** lfm-open-license-v1.0 (restricted)
- **Parameters:** 24.0B
- **Architecture:** MoE · 64 experts, 4/token · Lfm2MoeForCausalLM
- **Context:** 131,072 tokens (config.json)
- **Task:** text-generation · text
- **Signal:** 2 HF likes · 62 downloads
- **Archived:** 2026-11-14 (no longer listed by the collector; stats as last seen)

## Runs on

real quant files; sizing calculation, not tested.

- Q4 ~16.2 GB: 24GB GPU (3090 / 4090), 24GB GPU (7900 XTX), Mac 24GB, Mac 36GB, CPU only, 32GB RAM (slow)
- Q8 ~27.2 GB: 32GB GPU (5090), 48GB GPU (RTX 6000 Ada / A6000), Arc B70 32GB, Mac 48GB, Mac 64GB (M1 Max)
- FP16 ~49.6 GB: 96GB GPU (RTX Pro 6000 Blackwell), 1× DGX Spark (128GB), 2× DGX Spark (256GB), Strix Halo 128GB unified (~96GB usable), Mac 96GB, Mac 128GB, Mac 192GB, Mac 256GB, Mac 512GB, CPU only, 64GB RAM (slow), CPU only, 128GB RAM (slow), CPU only, 256GB RAM (slow)

## Links

- [Hugging Face](https://huggingface.co/summerAIO/diffusion-lfm2.5-24b-a2b-blocktune)
- [Ollama](https://ollama.com/library/lfm2.5)