Releasesby GlyphImage and video
- GPT Image 2.5 models: Image generation now supports gpt-image-2.5-flare and gpt-image-2.5-sunburst. Both models accept up to 10 reference images, quality levels through max, 14 fixed aspect ratios plus auto, and 1K, 2K or 4K output.
Also shipped Sep 21, 2026ElevenLabs in September 2026 →
BigQuery Studio query results pane shows recent run history
The Query results pane now shows a short history of recent runs for a query, including multi-statement queries, without switching to the Job history tab. It is generally available.
The query history system table is now generally available
The system.query.history system table is now generally available. Use it to access account-wide records for queries run on SQL warehouses or serverless compute, including execution status, duration metrics, and query source. See Query history system table reference .
Cortex AI Function Evaluation for measuring quality (Public Preview)
Snowflake Cortex AI Function Evaluation is now available in public preview, enabling customers to measure the output quality, cost, and token usage of custom AI Functions and Cortex AI calls against labeled datasets.
Cortex AI Function Optimization for more efficient AI implementations (Public Preview)
Snowflake Cortex AI Function Optimization is now available in public preview, enabling customers to automatically search across prompts and models for improved implementations of custom AI Functions.
Cortex Agents Compact API (Preview)
The Cortex Agents Compact API is now available in preview. Use the agent:compact endpoint to summarize a conversation and pass the compact representation to subsequent agent:run requests. Compacting conversation history reduces token consumption and helps conversations stay within the model context window.
You can now use the AlloyDB Columnar Engine as a read-optimized, in-memory cache for HNSW vector indexes
You can now use the AlloyDB Columnar Engine as a read-optimized, in-memory cache for HNSW vector indexes. This feature is generally available ( GA ). It accelerates vector search performance and increases queries per second (QPS) for vector workloads. For more information, see Accelerate queries with the Columnar Engine .
More from ElevenLabsEvery ElevenLabs release →
ElevenAgents
- Parallel tool calls: Agent prompt configuration now includes enable_parallel_tool_calls (boolean, default true). When enabled, supported models can execute multiple tools in one turn. Agent models: Added gpt-6-astra to the agent LLM options. Alerting: Agent alerting now supports Slack channels alongside PagerDuty and webhook notifiers.
Speech Engine
- Cascade timeout: Speech Engine create and update requests now accept cascade_timeout_seconds (number, 2–15, default 4) to control how long ElevenLabs waits for an upstream speech engine before retrying.
Text to Speech
- Multi-context turn boundaries: The is_final_audio_for_turn marker is now emitted after every buffered byte for that turn, including MP3 and Opus output. Clients can use the marker as an exact turn boundary without switching to PCM.
SDK releases: JavaScript SDK
- v2.69.0 - Added transcriptEdit to the realtime Speech to Text wrapper, with edited text delivered through edited_transcript events. Fixed Speech Engine JWT verification for API keys with data residency suffixes. Added phoneNumbers.listV2 and Flows template methods to list and get templates and list, create and get runs.
SDK releases: Python SDK
- v2.69.0 - Added transcript_edit to the realtime Speech to Text wrapper, with edited text delivered through edited_transcript events. Fixed Speech Engine JWT verification for API keys with data residency suffixes, and made play raise an error when ffplay fails.
SDK releases: ElevenLabs CLI
- v1.3.2 - Added optional --intent metadata to collect feedback on CLI usage. Added elevenlabs feedback missing-capability to report workflows that the CLI does not support.
Image & Video
- Sora 2 and Sora 2 Pro retired: OpenAI is discontinuing the Sora API on September 24, 2026. Both models have been removed from the Image & Video model picker and stop generating on that date. Existing generations remain available in your history.