EN

AI 产品动态

查看新模型发布、限免和订阅覆盖变化。

时间线

查看统计

第 1 / 2 页 · 本页 25 条

来源时间 · 北京时间

    1. 动态

      grok-imagine-image-quality retired

      Requests to `grok-imagine-image-quality` and `grok-imagine-image-pro` are now served by `grok-imagine-image-2.0-low`: `grok-imagine-image-2.0` with `quality` set to `low`, at a lower per-image price. No request changes are needed. `usage.cost_in_usd_ticks` on each response reflects the lower price, and gRPC responses (xAI SDK) report `grok-imagine-image-2.0-low` as the `model`. See the [migration guide](/developers/migration/imagine-image-quality-nov-2).

    2. 动态

      New Tools Available

      **Collections Search Tool**: You can now search through uploaded knowledge bases (collections) in chat conversations via the API. For more details, check out the [docs](/developers/tools/collections-search). **Remote MCP Tools**: You can now use tools from remote MCP servers in chat conversations via the API. For more details, check out the [docs](/developers/tools/remote-mcp). **Mixing client-side and server-side tools**: You can now mix client-side and server-side tools in the same chat conversation. For more details, check out the [docs](/developers/tools/advanced-usage#mixing-server-side-and-client-side-tools).

    1. 动态

      safety\_identifier request field

      You can now send `safety_identifier`, an opaque end-user identifier assigned by your application, on Chat Completions, the Responses API, deferred chat completions, the Batch API, and the gRPC `GetCompletionsRequest`. It lets SpaceXSI attribute a policy violation to one of your end users rather than to your API key. The legacy `user` field is still accepted. See the [Security FAQ](/developers/faq/security#how-do-i-identify-which-of-my-end-users-caused-a-policy-violation).

    1. 动态

      Image API updates

      **Auto quality.** The `quality` parameter on `grok-imagine-image-2.0` now accepts `auto`, and the default when `quality` is omitted has moved from `medium` to `auto`. Auto currently uses `low` for image generation and `medium` for image editing. Images are billed at the quality they are served at. Pass `low` or `medium` explicitly to pin a specific quality. See [Image Generation](/developers/model-capabilities/images/generation#quality). **Five reference images.** Image editing now accepts up to 5 source images per request (was 3). See [Multi-Image Editing](/developers/model-capabilities/images/multi-image-editing). **New aspect ratios.** Image generation and editing accept `21:9` (cinematic widescreen) and `5:2` (wide banners). See [Image Generation](/developers/model-capab

    1. 动态

      grok-imagine-video-1.5 modalities

      `grok-imagine-video-1.5` now supports text-to-video, image-to-video, and reference-to-video (including optional preset voices), with native 1080p for T2V and I2V. Text-to-video on this model runs as text-to-image then image-to-video under the hood. See [Video Generation](/developers/model-capabilities/video/generation), [Image-to-Video](/developers/model-capabilities/video/image-to-video), and [Reference-to-Video](/developers/model-capabilities/video/reference-to-video).

    1. 动态

      Public URLs and Files API ↔ image & video integration

      **Public URLs for Files** — turn any file in your [Files API](/developers/files) storage into a permanent, unauthenticated URL that anyone can open, embed, or share. Revocable at any time, or set an auto-expiry between 1 hour and 30 days. See the [Public URLs docs](/developers/files/public-urls). **Reference stored files as image & video inputs** — substitute `image_file_id`, `video_file_id`, or `reference_image_file_ids` for URL inputs across every image & video endpoint, with no need to re-upload bytes or make the file public. See [Files API integration for image & video requests](/developers/model-capabilities/imagine/files/inputs). **Persist image & video outputs to Files** — set `storage_options` on any image or video request to save the generated asset to your Files st

    1. 动态

      Smart Turn for Streaming STT

      The streaming Speech to Text API now supports Smart Turn end-of-turn detection. When enabled via the `smart_turn` query parameter, an ML model predicts whether the speaker has finished their thought at silence boundaries — reducing false endpointing during dictation, number sequences, and mid-sentence pauses. Use `smart_turn_timeout` to set a maximum silence fallback. For more details, see the [Smart Turn docs](/developers/model-capabilities/audio/speech-to-text#smart-turn).