Small Local Models + Fine-Tuning Instead of More Compute (Liquid AI, LEAP)

Melvin Vivas · X post · 2026-08-09 · Open on X

Topics: Fine-tuning & Model Customization, LLM Fundamentals, Industry Trends & Job Market · Level: intermediate

Summary

Melvin argues that instead of chasing more compute, you should make small local models work for your specific use case. He points to Liquid AI's small models, which can even run on phones, and to LEAP, Liquid AI's framework for fine-tuning a model for your use case.

Key points

• Don't chase more compute. Fit small local models to your specific use case instead.
• Liquid AI makes small models that are a good starting point.
• Liquid AI models are small enough to run on phones.
• LEAP is Liquid AI's fine-tuning framework for training a model for your use case.

Resources mentioned

• Liquid AI (@liquidai) on X · website · x.com · free
  An AI company that builds efficient foundation models. The quoted post shows its PII handling working on Japanese text.
  Also in: Zero-Shot Prompt Routing with Liquid AI's LFM 2.5-Encoder-350M (Melvin Vivas on X · notes), Liquid AI LFM 2.5 Encoder: a CPU-friendly encoder model (Melvin Vivas on X · notes), Fine-tuning Liquid AI LFM2/LFM2.5 MoE models with the new Halo framework (Melvin Vivas on X · notes), Liquid AI's LFM2-Longevity models for aging-data analysis (Melvin Vivas on X · notes) and 19 more
• LEAP (Liquid AI) · tool · leap.liquid.ai · check price
  Liquid AI's framework for fine-tuning and customizing its small models.
  Also in: Smoke-Testing LFM Fine-tuning with Codex and Liquid AI's LEAP (Melvin Vivas on X · notes), LFM2.5-2.6B Matches DeepSeek-V4-Flash on Tool Calling; LEAP Fine-Tuning (Melvin Vivas on X · notes)

Try this

• Try Liquid AI's small models for your own use case.
• Look into LEAP to fine-tune a small model for a specific task.
• Fine-tune a small Liquid AI model with LEAP for one specific use case and run it locally or on a phone.