New dedicated endpoint models
The following models are now available for deployment on dedicated endpoints : nvidia/nemotron-3-asr-streaming-0.6b (speech-to-text). nvidia/nemotron-3.5-asr-streaming-0.6b (speech-to-text).
Also shipped Nov 11, 2026Together AI in November 2026 →
Web search in apps
Retool added native web search capabilities to the app builder Chat tab and app functions, enabling agent applications to query live web data via a managed Tavily account.
Eval datasets, curated by Bayes from your agents' real conversations
Monte Carlo introduced automated evaluation dataset generation by Bayes, creating Golden and Improvement datasets from an agent's scored conversations across intent and issue clusters.
Pages are generally available
Databricks announced the general availability of Pages in Unity Catalog, providing governed definitions of business concepts organized by domain on the Discover page.
BigQuery project caps are generally available
Project caps, also known as scheduling policies, are now generally available for BigQuery Enterprise and Enterprise Plus editions. This feature allows users to limit maximum slots and concurrency per project within a reservation.
BigQuery reservation assignments support principal routing
BigQuery now allows specifying a principal property on reservation assignments to route queries to specific reservations based on the identity of the user or service account. This feature is generally available for Enterprise and Enterprise Plus editions.
BigQuery JSON_FLATTEN function is generally available
The JSON_FLATTEN function is now generally available. It extracts non-array values from JSON inputs, including those nested within consecutive arrays.
More from Together AI platformEvery Together AI platform release →
New serverless models
Together AI added thinkingmachines/Inkling-2-Small to its serverless API platform, offering a 1M token context length, NVFP4 quantization, and structured outputs.
Evaluations retired
Together AI has shut down its Evaluations service. This includes the removal of the evaluations API, associated SDK methods, CLI commands, and the console interface.
New serverless models
The following models are now available on serverless : elorian-labs/glimpse-1.0 : 262,144 context length, FP8 quantization, function calling, and structured outputs. Pricing: $0.60 input / $3.00 output / $0.06 cached input (per 1M tokens).
Post-training: Nemotron 3.5 Super VL
Together AI added post-training support for NVIDIA Nemotron 3.5 Super VL. The platform supports LoRA training on B200-SXM GPUs with context windows up to 163,840 tokens, rank up to 32, and up to three adapter sessions per resource.
New dedicated endpoint models
The following models are now available for deployment on dedicated endpoints : Qwen/Qwen3-TTS-12Hz-1.7B (text-to-speech). This model was previously listed as Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice . thinkingmachines/Inkling-2-Small .
Post-training: sampling-only model resources
A post-training model resource can now run generators with no trainer. To sample a saved inference checkpoint without training it further, set num_trainer_replicas=0 . You're billed only for the generator GPUs. See Sampling-only resources .
What's new:
- Rollout capacity budgets: tg endpoints rollout takes two flags that set a rollout's capacity budgets . --max-surge caps how many replicas each canary step adds to the target. --max-unavailable lets a rollout start from a DEGRADED source, or shut down source replicas to make room for the target.
Default expiry for Jig secrets
New Jig secrets now expire after 14 days unless you set a shorter lifetime with --ttl . Before a secret expires, the project's admins and the secret's creator get a daily email reminder. An expired secret is deleted, so new replicas can't read it. Existing replicas keep the value until they terminate.