Llama · AI Reasoning · Apple · Hugging Face
Run fast, private agents on-device with LFM2.5-24B-A2B
Compiled by KHAO Editorial — aggregated from 1 source. See llms.txt for citation guidance.
★ Tier-1 Source
LFM2.5-24B-A2B builds on LFM2-24B-A2B and keeps its architecture: 40 layers, 64 experts per MoE block with top-4 routing, and the same hybrid of gated short convolutions and GQA.
Key facts
- The port depends on your serving backend: llama.cpp and MLX use 8080, vLLM uses 8000, SGLang uses 30000, and LM Studio uses 1234
- LFM2.5-24B-A2B is the fastest of the compared MoEs, decoding 184 tokens/s on an M5 Max and 114 tokens/s on a Ryzen AI Max+ 395
- The team evaluated LFM2.5-24B-A2B against Qwen3.6-35B-A3B, Gemma-4-26B-A4B-it, and Nemotron-3.5-Lightning on 16 benchmarks covering tool use and agentic workflows, knowledge, instruction following, math
- LFM2.5-24B-A2B builds on LFM2-24B-A2B and keeps its architecture: 40 layers, 64 experts per MoE block with top-4 routing, and the same hybrid of gated short convolutions and GQA
Summary
Best-in-class agent: Competitive with Qwen3.6-35B-A3B, a model with ~50% more total and active parameters, on agentic and tool-use tasks, scoring 48.09 on LiveClaw versus 38.19. Reasoning at MoE speed: Always thinks before it answers. Efficient inference: 184 tok/s on an Apple M5 Max and up to 21K output tok/s on a single NVIDIA B300 GPU. Pre-training was scaled up from ~17T to ~33T tokens.