{"slug": "open-convenient-and-predictable-introducing-provisioned-throughput", "title": "Open, convenient and predictable: Introducing Provisioned Throughput", "summary": "Together AI launched Provisioned Throughput, a new inference service offering reserved capacity for open-weight models with token-based pricing and a 99% uptime SLA, available for MiniMax M3 and GLM-5.2. The service aims to provide predictable performance and cost savings up to 90% below proprietary alternatives, as companies shift from closed APIs to open models for production workloads.", "body_md": "We're excited to introduce Provisioned Throughput, reserved inference capacity for frontier open models with token-based pricing and a 99% uptime SLA.\n\nOpen weight models have become the essential ingredient for any company that wants an abundant AI future. But historically to use open weight models companies have had to choose between convenient but best-effort serverless or powerful but tunable dedicated inference. Provisioned Throughput is a new inference form factor that offers an alternate option that gives customers the simplicity of pre-optimized models at token prices with the predictability of guaranteed capacity in with an availability SLA. Costs run up to 90% below Claude Opus 4.8 at list price.\n\nAvailable today for MiniMax M3 and GLM-5.2, with capacity in North America, EMEA, and beyond, and a one-month minimum term.\n\nInference spend has become a line item the board asks about.\n\nIn a not too distant future every knowledge worker will have several agents working on their behalf, but companies are struggling to achieve this without astronomic inference budgets. To succeed companies need to consume a range of models that span the Pareto frontier, matching task difficulty with model quality (and cost). Open weight models occupy an ever-greater fraction of that frontier.\n\nAs companies move to a multi-model, multi-harness setup, open models are powering more and more of their everyday agents for: coding, finance, marketing, and workflow automation.\n\nCompanies that have made the switch report 6-20x lower inference costs on open models compared to proprietary alternatives. We see the shift in our own numbers: In nine months, token volume through Together AI's APIs has grown from 30 billion to more than 400 trillion tokens a month. A substantial share of that traffic is workloads that used to run on closed APIs.\n\n**What's been missing: Token capacity you can trust**\n\nAbundant AI requires more than open models on the Pareto frontier. Abundant AI also requires peace of mind: committed capacity, predictable (and influenceable) pricing, and an SLA customers can depend on. Until now, the open-model market offered two options, and neither matched that shape. Serverless inference is largely best-effort: fine for development, but unpredictable for production workloads. Dedicated inference (i.e. dedicated GPUs) gives you guarantees and full control of the serving environment, but many teams don't want to manage the complexity of a fully configurable inference platform. They want a dependable model provider behind a simple API, not GPU-hour math, capacity planning, and a serving stack to configure.\n\nWhat teams actually want is simple: The same token-based capacity agreement they already have with their closed-model provider, pointed at frontier open models. That's the product we built.\n\n**Introducing Provisioned Throughput**\n\nWhereas Together AI’s serverless inference is “pay for what you use” with a best effort SLA, Provisioned Throughput is “pay for the guaranteed capacity to use” with an uptime SLA.\n\nYou buy Provisioned Throughput Units (PTUs). Each PTU is a fixed slice of capacity: a guaranteed rate of tokens per minute for the model you're running, held exclusively for you, priced at $0.05 per PTU per minute. Provisioned Throughput accounts for variations in your traffic; input, cached input, and output tokens burn down PTUs differently allowing you to optimize spend for your traffic patterns. There's no GPU-hour math and no infrastructure to manage. The inference API for Provisioned Throughput, Dedicated Inference and Serverless Inference is identical.\n\nProvisioned Throughput is available today for **MiniMax M3** and **GLM-5.2**, with more models coming. We have capacity available in North America, EMEA, and beyond, with a one-month minimum term and discounts at higher levels of commitment.\n\n\"Our models are built for demanding production workloads and Together AI's Provisioned Throughput gives teams a way to run them with the reliability those workloads require. As teams shift volume off closed APIs, Together AI's infrastructure lets them adopt M3 at scale with commitments they can build a business on. That's exactly the kind of access we want for our models.\"\n\n— Linda Sheng, President of Global Business, MiniMax\n\n**The economics**\n\nA PTU covers three types of traffic: input tokens, cached input tokens, and output tokens, each drawing down your capacity at a different rate. On MiniMax M3, one PTU delivers 138,840 input tokens per minute, 694,200 cached input tokens per minute, 23,140 output tokens per minute, or any mix of the three.\n\nReal workloads have different shapes. Output-heavy traffic consumes capacity faster than cache-efficient traffic, but traffic shape never changes the SLA. It only changes how fast you consume your capacity.\n\nAt full utilization, a PTU on MiniMax M3 works out to roughly $0.36 per million input tokens and $2.16 per million output tokens, against $5 and $25 list on Claude Opus 4.8. Across three representative production profiles:\n\nThese are estimates. Actual PTU usage will vary with your traffic shape. [Run your own numbers in the pricing calculator](https://www.together.ai/pricing#provisioned-throughput)\n\n**Where it fits**\n\nServerless is the fastest way to start building on open models. Provisioned Throughput is for production workloads on standard open models that need guarantees. If you need a fine-tuned or custom model with deep control over the serving environment, Dedicated Inference is the right fit.\n\nIf you're running production traffic on a proprietary API that a frontier open model can handle, the migration path is now complete: frontier-quality models, token-based pricing you can plan around, and reliability guarantees your customers can depend on, at up to 90% lower cost.", "url": "https://wpnews.pro/news/open-convenient-and-predictable-introducing-provisioned-throughput", "canonical_source": "https://www.together.ai/blog/provisioned-throughput", "published_at": "2026-07-08 00:00:00+00:00", "updated_at": "2026-07-08 07:43:06.345617+00:00", "lang": "en", "topics": ["ai-infrastructure", "ai-products", "ai-tools", "large-language-models", "ai-startups"], "entities": ["Together AI", "MiniMax M3", "GLM-5.2", "Claude Opus 4.8"], "alternates": {"html": "https://wpnews.pro/news/open-convenient-and-predictable-introducing-provisioned-throughput", "markdown": "https://wpnews.pro/news/open-convenient-and-predictable-introducing-provisioned-throughput.md", "text": "https://wpnews.pro/news/open-convenient-and-predictable-introducing-provisioned-throughput.txt", "jsonld": "https://wpnews.pro/news/open-convenient-and-predictable-introducing-provisioned-throughput.jsonld"}}