Liquid AI Releases LFM2.5-VL-3B: A 3B Vision-Language Model That Reads Screens, Grounds Objects, and Calls Tools On-Device
Evolving story · 2 updatesLiquid AI's LFM2.5-VL-3B Vision-Language ModelTimeline →Liquid AI has launched LFM2.5-VL-3B, a 3.1 billion parameter vision-language model designed for on-device use. It demonstrates strong performance in screen reading and object grounding, and introduces tool calling capabilities.
- Liquid AI's LFM2.5-VL-3B is a 3.1B parameter vision-language model for on-device deployment.
- The model shows improved performance in screen reading and object grounding tasks.
- It introduces on-device tool calling capabilities, a new feature for the VL line.
- The model is efficient, fitting in 3GB and achieving 228 tokens/s on an Apple M5 Max.
Liquid AI has introduced LFM2.5-VL-3B, a compact 3.1 billion parameter vision-language model optimized for on-device execution. This model achieves notable performance gains, scoring 80.7 on the ScreenSpot-v2 benchmark for screen reading and significantly improving object grounding accuracy on RefCOCO from 57.1 to 87.9.
A key advancement is the integration of tool calling functionality, a first for Liquid AI's vision-language series. The ToolSandbox benchmark shows a substantial leap from 26.4 to 59.5 with this new capability. The model's efficiency is highlighted by its 3 GB size and a decoding speed of 228 tokens per second on an Apple M5 Max chip, making it suitable for edge devices.
This release signifies a step towards more capable and accessible on-device AI, enabling complex visual understanding and interaction without constant cloud connectivity.
Enables building on-device AI applications with advanced vision and tool interaction.
Facilitates privacy-preserving, low-latency AI features on edge devices.
Indicates progress in efficient, deployable multimodal AI models.
Provides a model for learning and experimenting with on-device multimodal AI.
Advances the capability of AI that can understand and interact with visual information locally.
LLMGemini 3.7 Flash lands with coding gains and undercuts its three-week-old predecessor's price by 50%
LLMGoogle AI Just Released Gemini 3.7 Flash: A Coding and Agent Model at $0.75/1M Input Tokens
LLMGoogle announces Gemini 3.7 Flash just three weeks after previous release
LLMSpaceXAI Releases Grok 4.6: A 500K-Context Frontier Model Tuned for Long-Running Agents, Coding, and Knowledge Work
LLMNVIDIA AI Releases Nemotron 3.5 Lightning: A 30B Open MoE with 3B Active Parameters, and NeMo Switchyard Model Router
China urged to avoid ‘us or them’ split with US over AI governance - South China Morning Post
China is urged to avoid a 'us or them' split with the US over AI governance, amid growing tensions between the two nations.
Development of artificial intelligence could surpass China’s industrial boom, say Capital Group - The Corner .eu
Capital Group predicts AI development could exceed China's industrial boom, The Corner reports. This forecast highlights the potential for AI to drive significant economic growth.
HardwareOrganic-looking brake assemblies debut on new Czinger 21C Spyder
Czinger’s 21C Spyder hypercar features brake assemblies produced using topological optimization and additive manufacturing, marking a first for automotive braking systems.
Naabik'iyati' Committee approves the establishment of the Navajo Nation Artificial Intelligence Policy Working Group WINDOW ROCK, Ariz. – The Naabik'iyati' Committee approved a position statement and the creation of a Navajo Nation Artificial Intelligence P - facebook.com
The Naabik'iyati' Committee has approved the creation of a Navajo Nation Artificial Intelligence Policy Working Group. This group will develop a position statement on AI policy for the Navajo Nation.
AI ToolsMeet Needle 2: An Open 45M-Parameter Tool-Calling Model That Ships as a 14MB Binary and Runs a Full Session in 28MB of RAM
Cactus Compute has released Needle 2, a 45M-parameter AI model that runs efficiently on low-end hardware.
As Duke leans into AI, here are the free tools the University offers - The Duke Chronicle
Duke University is making its AI tools available for free to the public, as part of its efforts to lean into AI research and development.