LiquidAI: LFM2.5-2.6B

by Liquidtext->text1 endpoint
Neura Intelligence Index
9.6/ 100
Rank
#348—
Confidence
low
Benchmarks
1(3% coverage)

Overview

LFM2.5-2.6B is a compact reasoning model from Liquid AI. It is suited for agent workflows, data extraction, RAG, and long-context processing. Liquid advises against using it for agentic coding or knowledge-heavy tasks.

Prompts and outputs may be retained and used to train Liquid models.

Capabilities

Text generation
Tool use / Function calling
Structured output
Extended reasoning
Prompt caching
Long context
Top-P sampling
Stop sequences
Deterministic seed
Log probabilities

Modalities

Input
Text
Output
Text

Technical Specifications

Context Window
128K tokens
Max Output
8.2K tokens
Tokenizer
Other
Uptime (24h)
10000.0%
Parameters
2690000000B
License
lfm1_0Open Weights
Country
🇺🇸US
Release Date
August 4, 20263mo

Supported Parameters

frequency_penaltyinclude_reasoninglogprobsmax_completion_tokensmax_tokensmin_ppresence_penaltyreasoningrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p

Benchmark Performance

math

BenchmarkScoreSource
AIME 2025
51.9
verified

Performance Over Time

Pricing

Per 1M tokens
Input$0.35
Output$0.50
Cache read$0.04
Added
August 11, 2026
Last synced 11/5/2026