Together AI at NVIDIA GTC 2026: Dynamo, Multi-Agent Models and Voice AI
Together AI arrived at NVIDIA GTC 2026 with four major announcements. The company integrated NVIDIA Dynamo 1.0 into its inference stack and, together with NVIDIA, launched NemoClaw, an open-source tool that deploys AI agents with a single command. The Together platform now also offers Nemotron 3 Super, a hybrid 120B-parameter model with a 1M-token context window, and Parakeet TDT 0.6B V3, a speech recognition (ASR) model for real-time transcription.
AI-processed from Together AI Blog; edited by Hamidun News
Together AI came to NVIDIA GTC 2026 in San Jose with four major announcements, ranging from a new inference engine to voice AI and a powerful model for multi-agent workflows. The conference's central theme matches the company's strategic course: AI systems are becoming more open, more agentic and ready for production use. For developers and AI teams, this means easier access to tools that previously demanded significant resources.
Dynamo 1.0 and agentic infrastructure
NVIDIA released Dynamo 1.0, an open software engine for generative and agentic inference at production scale. Together AI has already integrated Dynamo into its inference stack and is actively using it to optimize real workloads. According to the company, this delivers higher performance on heavy requests at lower cost.
At the same time, NVIDIA and Together AI jointly launched NVIDIA NemoClaw, an open-source stack that simplifies deploying always-on AI assistants. With a single command, it installs NVIDIA OpenShell, a secure environment for running autonomous agents, and opens access to open models, including NVIDIA Nemotron. For developers building agentic systems, this means direct access to Together's library of 150+ optimized models, with dedicated endpoints built for production loads. Combining NemoClaw with Together's dedicated infrastructure lowers the barrier for teams that want to launch agentic AI products without lengthy setup.
Nemotron 3 Super: complex reasoning and multi-agent tasks
NVIDIA Nemotron 3 Super is a hybrid mixture-of-experts model built on a Mamba-Transformer architecture. It is designed specifically for complex tasks that require long-horizon reasoning and for scenarios in which several agents interact within a single workflow.
Key characteristics of the model:
- 120B parameters in total, with only 12B active per token, which substantially reduces the compute load
- A 1M-token context window for long-horizon reasoning tasks
- Optimization for running several agents in parallel, even on a single GPU
- Use cases: software development, financial analysis, cybersecurity automation
The model is available through Together AI's Dedicated Model Inference. Developers get a simple, scalable way to run advanced reasoning models in production without having to build their own infrastructure from scratch.
Voice AI: Parakeet for real-time transcription
A separate announcement added NVIDIA Parakeet TDT 0.6B V3 to Together AI's model library. It is a low-latency ASR (automatic speech recognition) model optimized for real-time applications. Parakeet offers high transcription accuracy along with the performance that conversational AI agents require.
Paired with Together's high-performance inference infrastructure, developers get a ready-made stack for building voice agents, from accurate transcription to scalable request serving. Potential uses include voice interfaces in customer support, healthcare, education and corporate communications, where recognition speed and reliability are critical.
"AI systems are becoming more open, more agentic and ready for production" — Together AI on the main theme of GTC 2026.
At the conference, the Together AI team also held technical sessions with customers, including Cursor (an AI assistant for developers) and Decagon (customer support automation), demonstrating real-world uses of the platform in software development and business process automation.
What this means
Together AI is steadily strengthening its position as an "AI Native Cloud": a single platform where open models, agentic infrastructure and voice AI are available to developers in one place. Tight integration with the NVIDIA ecosystem through Dynamo, NemoClaw and Parakeet makes Together a real alternative to closed solutions for teams that value infrastructure flexibility, predictable costs and full control over the models they use.
- Nov 19, 2026GitLab teaches AI agents to work on a schedule: from every 15 minutes to once a month
- Nov 19, 2026AWS AgentCore Gets Lambda Hooks and Uses 34% Fewer Tokens Than Alternatives
- Nov 19, 2026Open-Source Rust Video Editor FilmCraft Hits 18.5K GitHub Stars and Lets AI Agents Drive the Engine
- Nov 19, 2026Gemini API Can Now Load Tool Schemas on Demand Instead of in Every Prompt
Key Questions Answered
What are the key details on Dynamo 1.0 and agentic infrastructure?
NVIDIA released Dynamo 1.0, an open software engine for generative and agentic inference at production scale. Together AI has already integrated Dynamo into its inference stack and is actively using it to optimize real workloads. According to the company, this delivers higher performance on heavy requests at lower cost. At the same time, NVIDIA and Together AI jointly launched NVIDIA NemoClaw, an open-source stack that simplifies deploying always-on AI assistants.
What are the key details on Nemotron 3 Super: complex reasoning and multi-agent tasks?
NVIDIA Nemotron 3 Super is a hybrid mixture-of-experts model built on a Mamba-Transformer architecture. It is designed specifically for complex tasks that require long-horizon reasoning and for scenarios in which several agents interact within a single workflow.
What are the key details on Voice AI: Parakeet for real-time transcription?
A separate announcement added NVIDIA Parakeet TDT 0.6B V3 to Together AI's model library. It is a low-latency ASR (automatic speech recognition) model optimized for real-time applications. Parakeet offers high transcription accuracy along with the performance that conversational AI agents require. Paired with Together's high-performance inference infrastructure, developers get a ready-made stack for building voice agents, from accurate transcription to scalable request serving.
Test Frontier LLMs & Autonomous Agents in One Workspace
Run Claude Opus 5.5, GPT-6, Gemini 3.1, and DeepSeek side-by-side in Nomad AI (nomadnet.ai) — unified model routing, deep research, and multimodal workflows under one subscription.
The AI world, distilled — once a week
Seven stories that actually mattered, hand-picked. No noise, no reposts, no press releases.
Done! Check your inbox for a confirmation.