When was LFM2.5-2.6B released?

LFM2.5-2.6B was released on August 4, 2026 by Liquid AI. This is the official LFM2.5-2.6B release date tracked on LLM Stats.

How big is LFM2.5-2.6B?

LFM2.5-2.6B has 2.7 billion parameters. It was trained on 34.0 trillion tokens. It ships as an open-weight model, so you can download and run it on your own hardware.

Who created LFM2.5-2.6B?

LFM2.5-2.6B was created by Liquid AI.

What is the license for LFM2.5-2.6B?

LFM2.5-2.6B is released under the LFM Open License v1.0 license. This is an open-source / open-weight license that permits self-hosting.

Where is the LFM2.5-2.6B paper or technical report?

LFM2.5-2.6B has a paper or technical report available at https://arxiv.org/abs/2511.23404. Use that source for architecture, training, release and evaluation details.

What models should I compare LFM2.5-2.6B against?

Common LFM2.5-2.6B comparisons include LFM2.5-2.6B vs Claude 3.7 Sonnet, LFM2.5-2.6B vs Qwen3 VL 32B Thinking, LFM2.5-2.6B vs Qwen3 VL 30B A3B Thinking. Compare them side by side for benchmark scores, pricing, context window, latency and API availability.

LFM2.5-2.6B API Pricing, Context Window & Benchmarks

Name: LFM2.5-2.6B
Author: Liquid AI

LFM2.5-2.6B: API Pricing, Context Window & Benchmarks

LFM2.5-2.6B is a language model from Liquid AI, released in August 2026.

LFM2.5-2.6B is Liquid AI's open-weight 2.6B dense hybrid text→text model (LFM2 architecture) post-trained for on-device agentic workloads: planning, native tool calling, and multi-step tasks. ~2.69B parameters, trained on ~34T tokens, 128K

LFM2.5-2.6B benchmarks

Capability tiers

Standing within each category, adjusted for leaderboard depth.

Real tasks performance

High-confidence performance for LFM2.5-2.6B across real-world prompt categories. Only 95% intervals at most 4 points wide are shown.

Performance by conversation depth

How LFM2.5-2.6B holds up as conversations get longer.

Quality Tracker

LFM2.5-2.6B Performance Across Datasets

Scores sourced from the model's scorecard, paper, or official blog posts

llm-stats.com - Mon Sep 07 2026

Notice missing or incorrect data?

LFM2.5-2.6B model size

LFM2.5-2.6B has 2.7 billion parameters and was trained on 34 trillion tokens. See how it compares to other models in the same parameter range.

ParametersTraining tokens

2.7B

34Ttokens

12639× tokens-to-params ratio

Small (<3B)

2.7B

1B7B70B405B

LFM2.5-2.6B latency

LFM2.5-2.6B time to first token, sustained output throughput, and failed-request rate from live API traffic over the trailing 7 days.

LFM2.5-2.6B examples

Recent arena outputs from LFM2.5-2.6B, picked from the highest-ranked matchups.

LFM2.5-2.6B license

LFM2.5-2.6B is released under the LFM Open License v1.0 license, which permits commercial use, has 2.7B parameters.

License: LFM Open License v1.0; Commercial use allowed
Parameters: 2.7B

Liquid AI LFM Open License v1.0 — open weights with a commercial-use threshold (annual revenue $10M+ requires a separate commercial license; qualified non-profits exempt for non-commercial/research use).

LFM2.5-2.6B resources

Official sources for LFM2.5-2.6B: paper or system card, official launch post, source repository, model weights.

LFM2.5-2.6B vs other models

The most-compared alternatives to LFM2.5-2.6B are Claude 3.7 Sonnet, Qwen3 VL 32B Thinking, Qwen3 VL 30B A3B Thinking. Open any pair side-by-side for benchmarks, pricing, context, and latency.