Top stories

Live feed of AI lab newsrooms. First-party announcements, traced to the source.

Feed Chatter About Privacy Connect RSS

© 2026 forck.live

Top stories

Top stories

Liquid AI — forck.live

The last 8 weeks

10
postsFirst-partypeak 4
1
outlet reportedOn 1 postpeak 1
No. 24
by popularitysame as last week

|

|

Announcements Chatter

Liquid AI

Announcements Chatter Models The week Connect your AI

Announcements Chatter

CompaniesApple Runway Google Research GitHub Amazon NVIDIA OpenAI Liquid AI

The last 8 weeks

10
postsFirst-partypeak 4
1
outlet reportedOn 1 postpeak 1
No. 24
by popularitysame as last week

Lead story

Liquid AI· 1 day ago

Open d1: Edge decision models for text, vision, and audio

Open-weight decision models for edge and GPU inference

Open-weight release

Every post, as it lands

1 day agoOpen d1: Edge decision models for text, vision, and audio
3 days agoIntroducing d1: The most capable decision model, now with vision
14 days agoLFM2.5-VL-DSpark: Accelerating vision-language models on edge and beyond
15 days agoLiquid AI, in Collaboration with Qualcomm Technologies, Brings Personal AI Context to Devices Powered by Snapdragon
21 days agoSetting the Frontier of Aging Biology with Liquid Foundation Models
24 AugIntroducing Pipette: A benchmarking suite for on-device intelligence
20 AugUp to 3.2x Faster Inference with LFM2.5-DSpark
20 AugLFM2.5-DSpark: Up to 3.2x Faster Inference from H100 to MacBook
19 AugLFM2.5 Q4_0: Quantization-Aware Distillation for Edge Deployment
18 AugDesigning Loops for Production-Grade Work
12 AugLFM2.5-VL-3B for Better and Faster Vision Capabilities for the Edge
12 AugLFM2.5-VL-3B: A Better and Faster Vision-Language Model for the Edge

Everything else, by date

Introducing d1: The most capable decision model, now with vision5 Oct
LFM2.5-VL-DSpark: Accelerating vision-language models on edge and beyondModel update24 Sep
Liquid AI, in Collaboration with Qualcomm Technologies, Brings Personal AI Context to Devices Powered by SnapdragonCapability change23 Sep
Setting the Frontier of Aging Biology with Liquid Foundation Models17 Sep
Introducing Pipette: A benchmarking suite for on-device intelligence24 Aug
Up to 3.2x Faster Inference with LFM2.5-DSparkInfrastructure20 Aug
LFM2.5-DSpark: Up to 3.2x Faster Inference from H100 to MacBook20 Aug
LFM2.5 Q4_0: Quantization-Aware Distillation for Edge Deployment19 Aug
Designing Loops for Production-Grade Work18 Aug
LFM2.5-VL-3B for Better and Faster Vision Capabilities for the Edge12 Aug
LFM2.5-VL-3B: A Better and Faster Vision-Language Model for the Edge12 Aug
MacPaw Partners with Liquid AI to Bring On-Device AI to Millions of Mac UsersPartnership5 Aug

+11 more postsShow fewer

LFM2.5-2.6B: Deploy Agents EverywhereNew model4 Aug
LFM2.5-Encoders for Fast Long-Context Inference on CPUOpen-weight release28 Jul
LFM2.5-Encoders: Fast at Long Context, Even on CPU28 Jul
Tokenizer Expansion: Upgrading a Model's Tokenizer in Place

Ask your AI

Your brief, straight from the record

Connect once, then ask Claude or ChatGPT what changed for the tools you use. Every item is cited to its record.

+ any MCP app

Connect your AI →

Every post, as it lands

1 day agoOpen d1: Edge decision models for text, vision, and audio
3 days agoIntroducing d1: The most capable decision model, now with vision
14 days agoLFM2.5-VL-DSpark: Accelerating vision-language models on edge and beyond
15 days agoLiquid AI, in Collaboration with Qualcomm Technologies, Brings Personal AI Context to Devices Powered by Snapdragon
21 days agoSetting the Frontier of Aging Biology with Liquid Foundation Models
24 AugIntroducing Pipette: A benchmarking suite for on-device intelligence
20 AugUp to 3.2x Faster Inference with LFM2.5-DSpark
20 AugLFM2.5-DSpark: Up to 3.2x Faster Inference from H100 to MacBook
19 AugLFM2.5 Q4_0: Quantization-Aware Distillation for Edge Deployment
18 AugDesigning Loops for Production-Grade Work
12 AugLFM2.5-VL-3B for Better and Faster Vision Capabilities for the Edge
12 AugLFM2.5-VL-3B: A Better and Faster Vision-Language Model for the Edge

Ask your AI

Your brief, straight from the record

Connect once, then ask Claude or ChatGPT what changed for the tools you use. Every item is cited to its record.

+ any MCP app

Connect your AI →

21 Jul

Reducing Doom Loops with Final Token Preference Optimization7 Jul

IFStruct: Measuring structured-output compliance30 Jun

LFM2.5-230M: Built to Run Anywhere25 Jun

LFM2.5 Retrievers: Bi-directional LFMs for Fast Multilingual Search18 Jun

LFM2.5-8B-A1B: An Even Better On-Device Mixture of Experts28 May

LFM2.5-VL-450M: Structured Visual Intelligence, Edge to Cloud8 Apr

LFM2.5-350M: No Size Left Behind31 Mar