modelgrep

Best LiquidAI Models for Agents

Quick answer · Updated June 2026

LFM2.5-1.2B-Thinking (free) is the best LiquidAI model for agents, scoring 6.5 on the Artificial Analysis Agentic Index for tool use and multi-step task completion. LFM2-24B-A2B (3.7) and LFM2.5-1.2B-Instruct (free) (3.6) round out the top three.

6.5Agentic
8.1Intelligence
40 t/sSpeed
FreeInput /M
33KContext

AI models ranked by the Artificial Analysis Agentic Index — measuring multi-step tool use, planning and task completion (including Tau²-Bench). The best models for building autonomous agents and agentic workflows.

  1. 1L
    lfm-2.5-1.2b-thinking:free
    Reasoning8.1 intel · Free/M · 40 t/s
    6.5
    Agentic
  2. 2L
    lfm-2-24b-a2b
    10.5 intel · $0.030/M · 63 t/s
    3.7
    Agentic
  3. 3L
    lfm-2.5-1.2b-instruct:free
    8.0 intel · Free/M · 94 t/s
    3.6
    Agentic

Frequently asked

What is the best LiquidAI model for agents?

LFM2.5-1.2B-Thinking (free) is the best LiquidAI model for agents, scoring 6.5 on the Artificial Analysis Agentic Index for tool use and multi-step task completion. LFM2-24B-A2B (3.7) and LFM2.5-1.2B-Instruct (free) (3.6) round out the top three.

What's a good alternative to LFM2.5-1.2B-Thinking (free)?

LFM2-24B-A2B (3.7) is the closest alternative on this metric, followed by LFM2.5-1.2B-Instruct (free) (3.6). See the full ranking above for the tradeoffs.

How many LiquidAI models are there?

modelgrep tracks 3 LiquidAI models with live benchmarks, speed, latency and per-provider pricing, led on intelligence by LFM2-24B-A2B. 3 of them qualify for this ranking.

More LiquidAI rankings

All rankings