Liquid AI

company

https://www.liquid.ai

AI & ML interests

A new generation of foundation models from first principles.

Recent Activity

nathanrchn authored a paper 24 days ago

Apertus 1.5: From Open Text Models to Open Multimodal Reasoners

Aurelien-Lac authored a paper 2 months ago

NeoMME: A Single-Tower Multimodal-Native Multilingual Foundation Encoder for Efficient Fine-Tuning and Inference

mlabonne authored a paper 10 months ago

LFM2 Technical Report

View all activity

Papers

LFM2 Technical Report

View all Papers

Articles

Build specialized multimodal models with LFM3

Run fast, private agents on-device with LFM2.5-24B-A2B

Multimodal open d1 decision models for the edge

Accelerating vision-language models with LFM2.5-VL-DSpark

Up to 3.2x Faster Inference with LFM2.5-DSpark

LFM2.5 Q4\_0 Checkpoints from Quantization-Aware Distillation

LFM2.5-VL-3B for Better and Faster Vision Capabilities for the Edge

Deploy local agents everywhere with LFM2.5-2.6B

LFM2.5-Encoders for Fast Long-Context Inference on CPU

View all articles

LiquidAI 's models 245

LiquidAI/LFM2.5-1.2B-Thinking-MLX-5bit

Text Generation • 1B • Updated Aug 12 • 322 • 6

LiquidAI/LFM2.5-1.2B-Thinking-MLX-4bit

Text Generation • 1B • Updated Aug 12 • 823 • 12

LiquidAI/LFM2.5-350M-MLX-bf16

Text Generation • 0.4B • Updated Aug 12 • 343 • 8

LiquidAI/LFM2.5-350M-MLX-8bit

Text Generation • 0.4B • Updated Aug 12 • 1.35k • 14

LiquidAI/LFM2.5-350M-MLX-6bit

Text Generation • 0.4B • Updated Aug 12 • 331 • 11

LiquidAI/LFM2.5-350M-MLX-5bit

Text Generation • 0.4B • Updated Aug 12 • 311 • 9

LiquidAI/LFM2.5-350M-MLX-4bit

Text Generation • 0.4B • Updated Aug 12 • 788 • 13

LiquidAI/LFM2.5-230M-MLX-bf16

Text Generation • 0.2B • Updated Aug 12 • 374 • 7

LiquidAI/LFM2.5-230M-MLX-8bit

Text Generation • 0.2B • Updated Aug 12 • 412 • 9

LiquidAI/LFM2.5-230M-MLX-6bit

Text Generation • 0.2B • Updated Aug 12 • 304 • 6

LiquidAI/LFM2.5-230M-MLX-5bit

Text Generation • 0.2B • Updated Aug 12 • 300 • 6

LiquidAI/LFM2.5-230M-MLX-4bit

Text Generation • 0.2B • Updated Aug 12 • 1.53k • 6

LiquidAI/LFM2.5-VL-450M-MLX-bf16

Image-Text-to-Text • 0.4B • Updated Aug 12 • 311 • 11

LiquidAI/LFM2.5-VL-450M-MLX-8bit

Image-Text-to-Text • 0.4B • Updated Aug 12 • 370 • 10

LiquidAI/LFM2.5-VL-450M-MLX-6bit

Image-Text-to-Text • 0.4B • Updated Aug 12 • 305 • 9

LiquidAI/LFM2.5-VL-450M-MLX-5bit

Image-Text-to-Text • 0.4B • Updated Aug 12 • 289 • 8

LiquidAI/LFM2.5-VL-3B-MLX-bf16

Image-Text-to-Text • 3B • Updated Aug 12 • 411 • 16

LiquidAI/LFM2.5-VL-3B-MLX-8bit

Image-Text-to-Text • 3B • Updated Aug 12 • 690 • 18

LiquidAI/LFM2.5-VL-3B-MLX-6bit

Image-Text-to-Text • 3B • Updated Aug 12 • 380 • 14

LiquidAI/LFM2.5-VL-3B-MLX-5bit

Image-Text-to-Text • 3B • Updated Aug 12 • 381 • 14

LiquidAI/LFM2.5-VL-3B-MLX-4bit

Image-Text-to-Text • 3B • Updated Aug 12 • 4.31k • 17

LiquidAI/LFM2.5-2.6B-MLX-nvfp4

Text Generation • 3B • Updated Aug 12 • 420 • 4

LiquidAI/LFM2.5-2.6B-MLX-mxfp8

Text Generation • 3B • Updated Aug 12 • 376 • 3

LiquidAI/LFM2.5-2.6B-MLX-mxfp4

Text Generation • 3B • Updated Aug 12 • 378 • 4

LiquidAI/LFM2.5-2.6B-MLX-bf16

Text Generation • 3B • Updated Aug 12 • 484 • 7

LiquidAI/LFM2.5-2.6B-MLX-8bit

Text Generation • 3B • Updated Aug 12 • 6.23k • 3

LiquidAI/LFM2.5-2.6B-MLX-6bit

Text Generation • 3B • Updated Aug 12 • 476 • 4

LiquidAI/LFM2.5-2.6B-MLX-5bit

Text Generation • 3B • Updated Aug 12 • 398 • 3

LiquidAI/LFM2.5-2.6B-MLX-4bit

Text Generation • 3B • Updated Aug 12 • 3.43k • 4

LiquidAI/LFM2.5-2.6B-MLX

Text Generation • Updated Aug 6 • 27