Tencent Hy4 preview uses a recursive self-improvement loop that boosted its own inference throughput by 31.8%. The ...

Glimmer stakes out different ground: a dense model with native vision input, trained end-to-end around the agent loop, shipping with its own quantized variants and speculative-decoding drafter.

Two weeks of model cards and a day of throwing broken tools at it later, it's the first open model I've seen that's built ...

Nemotron 3.5 Lightning is designed for specialized tasks within multi-agent systems, while larger reasoning models can handle planning and orchestration. NVIDIA says it delivers up to 4x faster output ...

Earlier this week, the AI startup Liquid, formed in by former MIT computer scientists, debuted LFM2.5-2.6B, a new open-weight language model designed specifically for agentic work ...

Compare Mac Mini M6, M5 Pro, Mac Studio, RTX 5090, and RTX 4090 for running AI models locally. See which hardware runs ...

Meta has launched Muse Glimmer 30B, an open-weight AI model built for local agents, coding, tool use and multimodal tasks. The model posts strong benchmark scores against Gemma4-31B and Qwen3.6-27B ...

Nvidia Groq 3 LPX is now in full production, the first commercial-scale SRAM-based decode accelerator to ship, with Nebius ...

Researchers from UC Berkeley and MIT have developed FreeToken, an open-source inference engine that enhances the utility of ...

AI Coding Tip 030 - Turn Repeatable Skill Steps Into Tested Scripts Instead of Prompts 20 August —OpenCode announces the ...

Cerebras Systems (NASDAQ:CBRS) outlined a product roadmap centered on faster AI inference, expanded data-center capacity and ...

The healthcare industry in Texas continues to expand. As the population ages, the need for medical services grows. Increasing medical services means higher demand for medical billers and coders who ...