Google has introduced agent-based video analysis to its Gemini 3.7 Flash, 3.6 Flash, and 3.5 Flash-Lite models, allowing the AI to selectively examine video segments and adjust resolution, which the c…
Google has introduced agentic video understanding in its latest Gemini models—Gemini 3.7 Flash, 3.6 Flash, and 3.5 Flash-Lite—enabling more accurate and cost-efficient video analysis, with token usage…
Inception released Mercury 2.5, a diffusion large language model (dLLM) that generates tokens in parallel, achieving 1,107 tokens/sec on standard GPUs and a 10+ point intelligence jump over Mercury 2,…
Google DeepMind has launched agentic video understanding for Gemini 3.7 Flash, 3.6 Flash, and 3.5 Flash-Lite via the Gemini API, claiming an 88% reduction in token usage on long-video benchmarks while…
Google launched agentic video understanding across Gemini 3.7 Flash, 3.6 Flash, and 3.5 Flash-Lite, reducing analysis costs by up to 66% and token consumption by up to 88% while improving accuracy by …
Inception released Mercury 2.5 Preview on August 31, 2026, claiming 1,107 tokens per second on standard GPUs, a 10-point intelligence jump over Mercury 2, and quality comparable to GPT-5.6 Luna at low…
Google added Gemini 3.7 Flash to AI Mode on Friday, August 14, 2026, as a selectable option for Google AI Pro and Ultra subscribers in English, rolling out globally, one day after the model's general …
Google's Gemini app has replaced 3.6 Flash with 3.7 Flash in the model picker across Android, iOS, web, and macOS, following the model's launch on August 13, 2026. The update brings improved reasoning…
Google announced in July 2026 a series of AI updates, including three new Gemini models (Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber) for developers, the Gemini Robotics ER 2 embodied reason…
Google released Gemini 3.6 Flash, Gemini 3.5 Flash-Lite, and a government-gated 3.5 Flash Cyber on July 21, followed by DeepMind's Gemini Robotics 2 on July 30, signaling a strategy to commoditize the…
Google has expanded its Gemini family with the release of Gemini 3.6 Flash, Gemini 3.5 Flash-Lite, and Gemini 3.5 Flash Cyber, alongside the new Gemini Robotics 2 family for embodied control. The Flas…
Google's Gemini 3.6 Flash and Gemini 3.5 Flash-Lite models ignore the temperature, top_p, and top_k sampling parameters, according to Google's current migration guide. The three parameters are depreca…
Google launched Lyria 3.5, its latest music generation model, on July 29, 2026, promising richer melodic structures, higher-quality lyrics, and more emotionally nuanced vocals. The model, rolling out …
Firebase AI Logic now supports Gemini 3.6 Flash and Gemini 3.5 Flash-Lite, models Google shipped on July 21, 2026, with lower pricing than the previous Gemini 3.5 Flash. Gemini 3.6 Flash cuts output-t…
Google's revenue rose 24% year-over-year in its latest quarter, but the company reported its first negative free cash flow since its IPO 20 years ago after spending $6 billion on physical assets, with…
Google's Gemini API Managed Agents now default to Gemini 3.6 Flash, introduce environment hooks for custom pre- and post-tool execution scripts, and offer free tier access, enabling a single API call …
OpenAI's analysis of over 800,000 work-related messages from ChatGPT Business users found that 43.5% of occupation-specific tasks are performed outside the user's primary role, indicating roles are ex…
Google announced three new Gemini models on July 21—3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash-Cyber in CodeMender—while revealing that pre-training for Gemini 4 has begun, prompting users on Reddit to …
A developer measured a 30x cost swing on Google's Gemini 3.6 Flash by adjusting the reasoning_effort dial, with the minimal setting cutting per-call costs 91–97% on most tasks while breaking multi-ste…
Google DeepMind on July 21 released Gemini 3.6 Flash, Gemini 3.5 Flash-Lite, and Gemini 3.5 Flash Cyber, with 3.6 Flash offering 17% fewer output tokens and pricing cut to $7.50 per million tokens, wh…