Edge Model LFM2.5 Matches DeepSeek-V4 in Tool Calling at 3.7x the Speed
maximelabonne · x · 2026-08-07
For simple tool calling, edge models now demonstrate impressive capabilities. Liquid AI's LFM2.5-2.6B recently matched DeepSeek-V4 level performance in practical testing.
Running on a single machine with 4x RTX 5090s, both models successfully completed a complex job requiring 35 tool calls. LFM2.5 achieved a speed of 366 tok/s, finishing in just 19 seconds, whereas DeepSeek-V4-Flash ran at 77 tok/s and took 70 seconds. This highlights the potential of purpose-built small edge models for agentic workflows, capable of running directly on phones.
Related event: Liquid AI 2.6B Model Matches DeepSeek in Tool Calling, 3.7x Faster(3 posts)→
More from coding & agent
- Benchmarking 7 Self-Hosted Memory Providers for AI Agents with 71K Turns — max_paperclips · 2026-08-10
- CourseLit: Open-Source LMS for Selling Courses and Digital Downloads — tom_doerr · 2026-08-10
- ClawWork: Open-Source Project Tests if AI Can Earn Its Own Salary — dr_cintas · 2026-08-10
- Sunil Pai Proposes Every Company Needs a 'Cassandra' Agent — threepointone · 2026-08-10
- Open-Source Claude Code Plugin Generates Goldman Sachs-Style Equity Research Reports — tom_doerr · 2026-08-10
- AI Sidekicks Yield 20% Efficiency; Background Agents Deliver 10x Value — vasuman · 2026-08-10