Together AI Dedicated Inference
The facts below are what the maker publishes about Together AI Dedicated Inference, checked and dated. Our editors have not written about it yet.
Plans and prices
From Together AI Dedicated Inference's own pricing page, checked 8 Oct 2026.
| Plan | Price | What it gives you |
|---|---|---|
| Dedicated Inference — NVIDIA HGX H100 (on-demand) | Priced on request | Single-tenant GPU instances · guaranteed performance (no sharing) · custom models · autoscaling and traffic spike handling · On-demand, pay as you go; billed per GPU per hour |
| Dedicated Inference — NVIDIA HGX B200 (on-demand) | Priced on request | Single-tenant GPU instances · guaranteed performance (no sharing) · custom models · autoscaling and traffic spike handling · On-demand, pay as you go; billed per GPU per hour |
| Dedicated Inference — reserved capacity | Priced on request | Reserved GPU capacity · pricing depends on hardware and configuration · Contact sales |
The facts
Each one read from Together AI Dedicated Inference's own pages, with the date it was checked.
What it does
- API compatibility
- Dedicated and serverless inference use the same OpenAI-compatible API, so a serverless prototype can move to a dedicated endpoint by changing the endpoint.checked 8 Oct 2026
- Autoscaling and routing
- The product page lists SLO-driven autoscaling based on time to first token, latency, and throughput, along with least-loaded, session-affinity, and prefix-cache-aware routing.checked 8 Oct 2026
- Best suited for
- Together lists predictable or steady traffic, latency-sensitive applications, and high-throughput production workloads as best suited to Dedicated Model Inference.checked 8 Oct 2026
- Compliance and data residency
- Together says customers can choose a deployment region for data residency and identifies SOC 2 Type II and ISO 27001 as supporting certifications.checked 8 Oct 2026
- Custom models
- Customers can deploy custom or fine-tuned models from Hugging Face or their own S3 storage through the UI or CLI.checked 8 Oct 2026
- Deployment
- Customers select a target model and hardware configuration to deploy a production-ready endpoint, and Together says it can be live in minutes without deep infrastructure expertise.checked 8 Oct 2026
- Pricing basis
- Dedicated Inference is billed per GPU-hour based on GPU type and replica count; the pricing page lists on-demand rates for HGX H100 and HGX B200, while reserved rates require contacting sales.checked 8 Oct 2026
- Rollouts
- Customers can test updated deployments with shadow traffic or A/B testing, and Together says it manages the rollout without taking the serving endpoint down.checked 8 Oct 2026
- Security and privacy
- Dedicated deployments use single-tenant, isolated GPUs; Together says traffic and data are not shared and that it does not train on customer prompts or model weights.checked 8 Oct 2026
- Service
- Dedicated inference runs a model on GPUs reserved exclusively for a customer's workload, providing consistent latency, predictable cost, and control over the model and hardware.checked 8 Oct 2026
- Support
- The product FAQ directs users to contact Together's team for help, and the product page offers a Talk to Sales option.checked 8 Oct 2026
The company
- Company mission
- Together AI describes its mission as designing a full-stack AI platform powered by systems research to help teams ship faster, scale reliably, and achieve superior unit economics.checked 8 Oct 2026
- Founded
- 2022checked 28 Sep 2026
- Headquarters
- San Francisco, California, United Stateschecked 28 Sep 2026
Alternatives to Together AI Dedicated Inference
- 1
Dify
Free planOpen sourceWebAPIExtensionLinux8.4From $590 a year - 2
Replit
Free planAndroidiOSWebLinux8.4From $18 a month - 3
Bolt.new
Free planAndroidiOSWeb7.6From $25 a month - 4
Cursor
Free planiOSWebAPILinux7.6From $40 a month - 5
Flowise
Free planWebAPILinuxmacOS7.2From $35 a month - 6
OpenHands
Free planAndroidiOSWebAPI7.2Free plan