រំលងទៅមាតិកា
← កំពូល 50 នៅលើគេហទំព័រ

What's new in Google AI Studio

Google AI Studio · Google · វ៉ិប #11

  1. មុខងារថ្មី · · ផ្លូវការ · Google AI blog

    Google Opens Gemini API Tool Retrieval in Preview to Cut Token Use for Agents

    Google has launched Tool Retrieval in preview in the Gemini Interactions API. Instead of sending every tool's full details with each request, an agent loads only the details it needs, which is meant to cut input tokens and keep tool choice accurate.

    • Developers add a tool_search tool and set defer_loading to true on tools they want to hold back.
    • The model sees each deferred tool's name and description, and loads the full parameter details only when it needs that tool.
    • Gemini runs the search on its servers, so no separate search index needs to be built or hosted.
    • Tool Retrieval also works with remote MCP servers, and developers can run their own search on the client side.
    អានបន្ថែម ប្រភព ↗
  2. មុខងារថ្មី · · ផ្លូវការ · Google AI blog

    Google Search Gets Major Upgrade With Gemini 4 Hydrogen

    Google is rolling out its Gemini 4 Hydrogen AI model to Search worldwide. It better understands complex questions and shows answers with tables and images.

    • It better understands complex, multi-part questions.
    • It runs several related searches at once and links to a wider range of websites.
    • Answers include more images, comparison tables, and structured information.
    • Answers are more detailed and accurate in all supported languages.
    អានបន្ថែម ប្រភព ↗
  3. ម៉ូដែលថ្មី · · ផ្លូវការ · Google Gemini products

    Google Unveils Fast, Low-Cost Gemini 4 Hydrogen Model

    Google DeepMind has released Gemini 4 Hydrogen, a fast and efficient AI model. It is built for everyday coding and multi-step agent tasks. It is available in Google AI Studio, the Gemini API, the Gemini app, and Google Search.

    • It is lighter and faster than the frontier models Carbon and Argon, which suits repetitive work.
    • It scores slightly higher than Argon at default settings and finishes tasks faster than similar models.
    • It costs $1.50 per million input tokens and $7.50 per million output tokens.
    • A new Orchestrator feature mixes models in one session, cutting quota use by 24% and speeding work by 19% in coding tasks.
    អានបន្ថែម ប្រភព ↗

AI អានការបង្ហោះផ្លូវការ និងសារព័ត៌មានពាណិជ្ជកម្ម ហើយសរសេរវាឡើងវិញជាពាក្យសាមញ្ញ។ ធាតុមួយចំនួនអាចនឹងបាត់ ឬខុស។ សូមពិនិត្យមើលប្រភពសម្រាប់ព័ត៌មានលម្អិត។

អ្នក