Liquid AI's Labonne: Stop Using Free LFM2.5-2.6B for Coding, It's Not Built for It
maximelabonne · x · 2026-09-15
Liquid AI's Maxime Labonne jokingly urged users to stop using LFM2.5-2.6B for coding, then said 'I'm kidding, break stuff with it for free.' The OpenRouter listing confirms why: the free 2.6B reasoning model (15.4B tokens processed) is suited for agent workflows, data extraction, RAG and long-context tasks, but Liquid explicitly advises against it for agentic coding or knowledge-heavy work. The family also includes free LFM2.5-Embedding-350M, the on-device 24B MoE LFM2-24B-A2B (2B active params, fits in 32GB RAM), and LFM2.5-1.2B-Thinking. Free-tier prompts and outputs may be used for training.
Related event: Liquid AI Warns: Don't Use LFM2.5-2.6B for Coding(2 posts)→
More from Models
- Resemble AI ships DETECT-World, a physics-based deepfake detector with 99.5% audio accuracy — AiBreakfast · 2026-09-15
- Grok 4.7 Misses Target Again; 2.5T-Parameter Grok 4.8 Finishes Training — eyishazyer · 2026-09-15
- Writers Report Gemini Flash 3.8 Suffers Severe Long-Context Rot Despite 1M Token Window — Quenty1 · 2026-09-15
- Ai2 Chair's 3 Predictions: Open-Source AI Goes from Ideology to Infrastructure in 18 Months — billhilf · 2026-09-15
- Surgical VLM Leaderboard: All Frontier Models Fall Far Short of Specialized Models — ddonoho · 2026-09-15
- Why Surgery Benchmarks Reward Fine-tuned Small Models While Math Benchmarks Don't — ddonoho · 2026-09-15