Releasesby GlyphAutomatic idle shutdown for dedicated deployments
Deployments can now stop themselves when they go unused. Set an inactivity timeout with --inactive-timeout (the inactiveTimeout field in the management API), and if the deployment serves no inference requests for that many minutes, it scales to zero replicas, releasing its hardware and stopping billing. See Automatic idle shutdown .
Also shipped Sep 16, 2026Together AI in September 2026 →
BigQuery migration lineage service enters preview
The migration lineage service visualizes data flow and connections in a source database to help plan a BigQuery data warehouse migration.
Cortex Agents object enhancements (General availability)
The following Cortex Agents object enhancements are now generally available:
Include and exclude per-user quota users by name (General availability)
Per-user quotas now accept individual user names, so you can add or remove specific users alongside the quota's tag scope. INCLUDE_USERS is new and takes user names directly, and EXCLUDE_USERS accepts a 'USER' argument alongside its existing 'TAG' form:
Weekly per-user quota limits (General availability)
Per-user quotas now support a weekly credit limit alongside the existing monthly and daily limits. Pass 'WEEKLY' as the cycle argument to SET_PER_USER_LIMIT :
Anthropic models will be removed from AWS GovCloud IL5 workspaces
On September 19, 2026, Anthropic models will be removed from Databricks on AWS GovCloud and AWS GovCloud DoD workspaces that have the DoD Impact Level 5 (IL5) compliance standard selected. After this date, IL5 workspaces can't create or use endpoints that serve Anthropic models.
Unity Gateway API and developer tools are generally available
The Unity Gateway API for managing model services, model provider services, and MCP services is now generally available, with create, read, update, list, and delete support across Databricks developer tools: Terraform provider : 1.132.0 and later Databricks CLI : v1.17.0 and later Python SDK : databricks-sdk 0.136.0 and later Go SDK : databricks-sdk-go...
More from Together AI platformEvery Together AI platform release →
Rollouts for dedicated model inference
Rollouts shift live traffic from one deployment to another under the same endpoint, without changing the endpoint URL. Pick a canary, blue-green, or rolling strategy to determine how traffic moves, and optionally gate a canary rollout on live metrics so it pauses automatically if the new deployment regresses.
Together CLI v2.34.0
Version 2.34.0 of the Together CLI and Python SDK carries the rollouts release above (the SDK surface is client.beta.endpoints.rollouts ) and also adds: HIPAA placement : tg beta endpoints deploy accepts --placement.hipaa to restrict a deployment to HIPAA-attested clusters. The Python SDK takes the same policy as compliance_policy on inline placement.
Model deprecations
The following models have been deprecated and are no longer available on serverless: openai/gpt-oss-20b . Recommended replacement: Qwen/Qwen3.5-9B . Supported by on-demand dedicated endpoints. google/gemma-4-31B-it . Recommended replacement: zai-org/GLM-5.3-Flash . Supported by on-demand dedicated endpoints. thinkingmachines/Inkling-Small .
New serverless models
The following models are now available on serverless : MiniMaxAI/MiniMax-H3 : Pricing: $0.1391/sec at 2k.
New serverless models
The following models are now available on serverless : deepseek-ai/DeepSeek-V4.1-Flash : 1,000,000 context length, FP8 quantization, function calling, and structured outputs. Pricing: $0.30 input / $1.20 output / $0.006 cached input (per 1M tokens).
Pricing update
The following models have lower pricing, effective September 22, 2026. All usage from that date forward is billed at the new rates (per 1M tokens): Qwen/Qwen3.7-Max : $2.50 → $1.50 (input), $7.50 → $4.50 (output). Qwen/Qwen3.8-Flash : $0.15 → $0.09 (input), $0.47 → $0.282 (output). See Serverless models for the full pricing catalog.
Preemptible compute for GPU clusters
Preemptible compute is now in public preview for Kubernetes GPU clusters. Alongside standard nodes, you can set a preemptible GPU target, and Together provisions toward it as spare capacity becomes available, at a flat discounted rate relative to on-demand.
What's new:
- Preemptible GPU targets: Set num_preemptible_gpus at cluster create or update from the console, CLI, or API. Together automatically provisions replacements toward the target after nodes are reclaimed.