Today we release LFM2.5-Encoder-230M and LFM2.5-Encoder-350M: bidirectional encoders that stay fast at long context, even on CPU. > LFM2.5-Encoder-230M: about 3.7x faster than ModernBERT-base on CPU at 8,192 tokens. Under 30s per forward pass, versus over a minute and a half. > LFM2.5-Encoder-350M: 4th of 14 models on GLUE,... SuperGLUE, and multilingual classification, behind only three larger models, one of them nearly 10x its size. 🧵show more

Two sizes in the same family: > LFM2.5-Encoder-350M: the higher-accuracy pick. > LFM2.5-Encoder-230M: smaller and faster, for tighter hardware or higher throughput. Both are pre-trained with a masked-language objective, so you fine-tune them for classification, token-level tasks, or retrieval. (2/6)

Built on the LFM2 hybrid backbone, adapted from causal decoders into bidirectional encoders: bidirectional attention, non-causal short convolutions, and a 30% mask rate. The backbone's cost grows slowly as inputs get longer. > Context: 8,192 tokens, about 13 to 15 pages in one forward pass. > Runs on CPU with no GPU, where the speed edge is largest. > On GPU, the same pattern holds at the longest inputs. (3/6)

See it in action. We built demos that run entirely on CPU: > Prompt routing: > Policy linting: > Spell checker: Bonus: we even run the encoder as a masked-diffusion chatbot. (4/6)

LFM2.5-Encoder-230M and LFM2.5-Encoder-350M are available now. > LFM2.5-Encoder-230M: > LFM2.5-Encoder-350M: > Blog: > Docs:

Will there be ONNX weights released? Can't wait to run it on the edge!

Great news! Excited to try it out. LFM2-2.6B was a perfect base model for our C3LM!

Great job Viviana ❤️

Sub-30 second forward passes on a CPU at 8k tokens is absolute magic. 😊

952.4 t/s 0.6gb vram for 350M

LFM2 5 Encoders deliver long context speed on CPU

LiquidAI just makes a great model. I recently built an iOS app, that runs the LFM 8b a1b at 50 tokens per second. Crazy, right?

Fast long context on CPUs could quietly unlock many practical deployments. Impressive work.

You guys are doing the most important work right now, and making AI applicable and useful, thanks so much!

awesome! @zach_nussbaum

the under 30s figure, what batch size is that at? single sample vs batched throughput is a totally different story for anything doing live inference at scale

@Presidentlin

wow, this is awesome

incredible work! we are deeply admiring the open source contributions of @liquidai as @koniclabs keep up team 🫡

Great job, I am looking for a simple practical application in office wotk

8k token context on CPU at that speed is a game changer for lightweight pipelines

Beating a model 10x its size on GLUE is a huge efficiency win.

question: when would I use this over something like sentence-BERT? for larger documents and if I want it on CPU?

cpu wins at batch size 1 often disappear once batching enters production

Generative Reasoning Protocol (GRP) v0.6

4th of 14 on GLUE/SuperGLUE while being roughly a third the size of most competitors is a legitimate benchmark result, but the more interesting number is the 3.7x CPU speedup at 8,192 tokens — that's the one that actually determines whether prompt routing runs cheaply enough to sit in front of every request instead of being a nice-to-have.

Are you planning to publish the training code for the t the zero shot dual tower head, and an official ONNX

You guys beat me to it. This is what I've been trying to train for meept (

great job!👏

looks pretty cool, but is this worth it to use for LLM routing? by like choosing the subset that a task is in, out of hundreds, and then later after its classified pick the model that fits all criterias / scores best?

Impressive efficiency gains for long-context encoder workloads

Bidirectional encoders staying fast on CPU at 8k tokens is quietly a bigger deal than most launches this week. Reranking is the part of RAG everyone still ships to a GPU box out of pure habit. Curious how much of that 3.7x survives int8 quantization. Did you benchmark on commodity hardware too?

Great work!

350M 跑 128K 上下文延迟多少？

Babe wake up liquid AI just dropped a new thalamus routing model

interesting!