Liquid AI (@liquidai) / X

Liquid AI

660 posts

Liquid AI

@liquidai

Build efficient general-purpose AI at every scale.

Cambridge, MA

Joined March 2023

Pinned
Liquid AI
@liquidai
May 28
Today, we're releasing LFM2.5-8B-A1B, a device-optimized model designed to power real-life applications on phones, laptops, PCs, robots, and fast & lightweight server-side use-cases. > 8B MoE, 1.5B active > Expanded 128K context > LFM2.5 flagship hybrid MoE architecture >
1.3M
Liquid AI reposted
Noctus
@noctus91
Jun 20
LFM2.5-Embedding-350M + ColBERT from @liquidai makes local AI retrieval fast and surprisingly accurate. 95% top result accuracy on code questions. Strong multilingual retrieval. Around 10ms latency on device. Combined with LFM2.5-8B-A1B, you get a fully local AI agent. one
Liquid AI
@liquidai
Jun 18
Introducing LFM2.5-Embedding-350M and LFM2.5-ColBERT-350M: two multilingual retrieval models built for ultra-fast and accurate search across 11 languages. > End-to-end retrieval latency as low as 1.5ms with our enterprise stack! 🚀 > Consistently best-in-class multilingual and
3.9K
Liquid AI
@liquidai
Jun 19
Storing too many tools in your context window increases latency and can lead to wrong tool selection. In this demo, we used LFM2.5-ColBERT-350M as a filter to only select the five most relevant tools among 151 options. It's fast and reliable, even without any specific
00:00
31K
Liquid AI reposted
JJ
@JosephJacks_
Jun 18
Wow.. seems @liquidai is the only American lab outperforming the Chinese on open weights models! 🌊 📲
Liquid AI
@liquidai
Jun 18
Introducing LFM2.5-Embedding-350M and LFM2.5-ColBERT-350M: two multilingual retrieval models built for ultra-fast and accurate search across 11 languages. > End-to-end retrieval latency as low as 1.5ms with our enterprise stack! 🚀 > Consistently best-in-class multilingual and
3.1K
Liquid AI reposted
Antoine Chaffin
@antoine_chaffin
Jun 18
Turning a decoder into an encoder and training a late interaction model with this backbone... Maybe the propaganda is working, after all Very cool to see new multilingual ColBERT models popping up! We might have something on this topic for you very soon... which might create
Liquid AI
@liquidai
Jun 18
Introducing LFM2.5-Embedding-350M and LFM2.5-ColBERT-350M: two multilingual retrieval models built for ultra-fast and accurate search across 11 languages. > End-to-end retrieval latency as low as 1.5ms with our enterprise stack! 🚀 > Consistently best-in-class multilingual and
5.8K
Liquid AI reposted
Mathias Lechner
@mlech26l
Jun 18
Turning a causal LM into a bidirectional model
Liquid AI
@liquidai
Jun 18
Introducing LFM2.5-Embedding-350M and LFM2.5-ColBERT-350M: two multilingual retrieval models built for ultra-fast and accurate search across 11 languages. > End-to-end retrieval latency as low as 1.5ms with our enterprise stack! 🚀 > Consistently best-in-class multilingual and
2.4K
Liquid AI reposted
Maxime Labonne
@maximelabonne
Jun 18
We patched LFM2.5-350M (pre-trained on 28T tokens) to transform a causal decoder into a bidirectional encoder. It worked extremely well: both Embedding and ColBERT models get best-in-class performance, especially for multi/cross-lingual tasks.
Liquid AI
@liquidai
Jun 18
Introducing LFM2.5-Embedding-350M and LFM2.5-ColBERT-350M: two multilingual retrieval models built for ultra-fast and accurate search across 11 languages. > End-to-end retrieval latency as low as 1.5ms with our enterprise stack! 🚀 > Consistently best-in-class multilingual and
9.4K
Liquid AI reposted
Ramin
@ramin_m_h
Jun 18
1.5ms end-to-end retrieval latency ✨
Liquid AI
@liquidai
Jun 18
Introducing LFM2.5-Embedding-350M and LFM2.5-ColBERT-350M: two multilingual retrieval models built for ultra-fast and accurate search across 11 languages. > End-to-end retrieval latency as low as 1.5ms with our enterprise stack! 🚀 > Consistently best-in-class multilingual and
4.4K
Liquid AI
@liquidai
Jun 18
Introducing LFM2.5-Embedding-350M and LFM2.5-ColBERT-350M: two multilingual retrieval models built for ultra-fast and accurate search across 11 languages. > End-to-end retrieval latency as low as 1.5ms with our enterprise stack! 🚀 > Consistently best-in-class multilingual and
105K
Liquid AI
@liquidai
Jun 18
Replying to @liquidai
Both are built on LFM2.5-350M-Base with bidirectional patches, the first full-context encoders in the LFM family. > Both run via llama.cpp GGUFs (CPUs, Laptops, edge devices) <10ms end-to-end query embedding latency. > Both fly on GPUs with our custom runtime <2ms end-to-end
6.8K
Liquid AI
@liquidai
Jun 18
Especially well-suited for short-context collections: product catalogs, FAQ, knowledge bases, support docs, and multilingual search use cases where speed and reliability matter. LFM2.5-Embedding-350M and LFM2.5-ColBERT-350M are available now. > Blog post:
LFM2.5 Retrievers: Bi-directional LFMs for Fast Multilingual Search | Liquid AI
From liquid.ai
2.5K
Liquid AI
@liquidai
Jun 17
Our CTO Mathias Lechner, @mlech26l, and Yuri Khrustalev, @ykhrust, member of our technical staff, break down why @onnxai has become one of the most important standards in AI - and what it means for deploying Liquid Foundation Models in the real world.
00:00
7.8K
Liquid AI
@liquidai
Jun 17
More with @mlech26l and other members of our team coming soon!
1K
Liquid AI
@liquidai
Jun 11
We are proud to announce that Ion Stoica (@istoica05) co-founder of @databricks, @anyscalecompute, and @arena, and UC Berkeley Professor of Computer Science, has joined Liquid AI as a strategic member of our Advisory Council. Ion will guide us on our growth journey as we build
19K
Liquid AI
@liquidai
Jun 9
Most multimodal systems need data that combines every modality together. Hard to get, expensive to build. Our CTO Mathias Lechner, @mlech26l, sits down with Saniya Karwa, @saniwhya, from our multimodal research team to talk about building a mode that handles text, audio, and
00:00
15K
Liquid AI
@liquidai
Jun 9
More with @mlech26l and the team next week.
2.9K
Liquid AI reposted
t.toda
@Trtd6Trtd
Jun 8
LFM2.5-Audio-1.5B-JP試した！このレベルがLocalで動くのすごい！自音声拾ってしまうので、イヤホンでしか試せないのだが、割り込みもイケる
00:00
7K