New serverless models
Together AI launched tencent/Hy4 on its serverless platform, providing 1M context support, NVFP4 quantization, reasoning capabilities, multimodal input, and structured outputs.
Also shipped Nov 19, 2026Together AI in November 2026 →
Gemini 3.6 Flash will be retired on November 19, 2026
Databricks announced that Google Gemini 3.6 Flash will be retired on November 19, 2026, after which calls will automatically redirect to Gemini 3.8 Flash.
Wait for pull request events with GitHub CLI
GitHub CLI gh v2.105.0 added preview support for gh event listen, enabling users and agents to subscribe to pull request events without polling.
Gemini 3.6 Flash is retired
Google Gemini 3.6 Flash was retired on November 19, 2026, with API calls automatically redirecting to Gemini 3.8 Flash at the same price.
Gemini 4 Hydrogen in CodeMender (Preview)
Google Cloud added preview support for Gemini 4 Hydrogen in CodeMender, accessible by supplying the --model gemini-4-hydrogen flag to cm subcommands.
Data Engineering Agent observability is generally available in BigQuery
BigQuery added general availability for Data Engineering Agent observability, allowing users to track request volume, model latency, token usage, and tool traces in Cloud Monitoring and Cloud Trace.
BigQuery introduces AI indexes in Preview
Google Cloud released AI indexes in BigQuery in Preview, enabling natural language search across columns using the AI.SEARCH function without manual embedding generation.
More from Together AI platformEvery Together AI platform release →
What's new:
- Tinker 0.39.0 compatibility: The Tinker-compatible client now builds on tinker 0.39.0, which the tinker extra installs. It follows Tinker 0.39's call styles, so inside a running event loop you can await the same calls you make in synchronous code. ServiceClient now works as a context manager that closes every training client it created.
New serverless models
Qwen/Qwen4-Max is now available on serverless. It supports a 1,000,000 context length, reasoning, function calling, and structured outputs.
New models available for fine-tuning
NVIDIA-Nemotron-3.5-Lightning-30B-A3B and NVIDIA-Nemotron-3.5-Super-VL-120B-A12B are now available for fine-tuning. Both models support function-calling and reasoning.
What's new:
Together AI has updated its model resource management to force cleanup of resources when context managers exit. Additionally, evaluation commands have been removed from the CLI following the service shutdown.
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.
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 .