LiquidAI: LFM2-2.6B

LiquidAI: LFM2-2.6B

liquid · Released Oct 20, 2025
20
Our Score

LFM2 is a new generation of hybrid models developed by Liquid AI, specifically designed for edge AI and on-device deployment. It sets a new standard in terms of quality, speed, and memory efficiency.

$0.01 / 1M Input Price
$0.02 / 1M Output Price
32,768 tokens Context Window
6B Parameters

Architecture

ModalityText → Text
TokenizerOther
Parameters6B

Performance Indices

Source: Artificial Analysis

8 Intelligence Index
1.4 Coding Index
7.2 Agentic Index
8.3 Math Index

Benchmark Scores

Evaluations

GPQA Diamond 30.6%
Graduate-level scientific reasoning
HLE 5.2%
Humanity's Last Exam
MMLU Pro 29.8%
Multi-task language understanding
LiveCodeBench 8.1%
Live coding evaluation
SciCode 2.5%
Scientific computing
AIME 2025 8.3%
Competition mathematics (2025)
IFBench 19.5%
Instruction following
TerminalBench Hard 0.8%
Agentic terminal tasks
τ²-Bench 13.5%
Conversational agent benchmark

Benchmark data from Artificial Analysis and Hugging Face

Model Information

OpenRouter ID liquid/lfm-2.2-6b
Providerliquid
Release Date October 20, 2025
Context Length32,768 tokens
Status Active

Pricing

Token Type Cost per 1M tokens Cost per 1K tokens
Input $0.01 $0.000010
Output $0.02 $0.000020

Live Performance

Live endpoint metrics — refreshed every 30 minutes.

100%
Avg Uptime
357ms
Best Latency (TTFT)
133.5 tok/s
Best Throughput
1/1
Active Endpoints
Available via: Liquid