Skip to content
Rush Commerce
AI & Automation3 min read

Together AI raises $800M — open-source inference just went mainstream

Together AI's $800M round at an $8.3B valuation, led by Aramco's Prosperity7, signals open-weight models are now a serious cost play for small businesses.

While everyone watched OpenAI and Anthropic trade headlines, the money quietly moved somewhere else. On July 1, Together AI raised $800 million at an $8.3 billion valuation — a round led by Aramco's Prosperity7 Ventures — to run open-weight models as a hosted service. For a small business, the number that matters isn't the valuation. It's what it tells you: the open-source alternative to the big closed APIs is now well-funded, fast, and cheap enough that serious investors are betting nine figures on it.

What actually happened

Together AI closed an $800 million Series C at an $8.3 billion valuation, led by Aramco Ventures' Prosperity7, with Vista Equity Partners, General Catalyst, Emergence Capital, NVIDIA, and others joining (per BusinessWire). The company reported annual bookings topping $1.15 billion as enterprises shift workloads toward open models. The pitch, in one investor's words: "the shift toward open source isn't a niche preference anymore, it's becoming the default for any company that wants to scale AI without losing its margin" (via Pulse 2.0).

Together AI hosts open-weight models — the kind you can also run yourself — behind a standard API, so you get frontier-adjacent quality without the frontier-flagship price or the lock-in.

Why it matters for your business

"Without losing its margin" is the whole story. Most business AI work — sorting tickets, drafting replies, extracting fields, summarizing calls — does not need the most expensive closed model on the market. It needs a competent model at a price that survives contact with your P&L. Open-weight models running on a host like Together are increasingly that model.

The strategic win is portability. When your automation calls an open model behind a standard API, you are not married to one vendor's pricing whims. You can move that same workload to another host, or onto your own hardware, without rewriting the logic. That's the opposite of building your business on a closed flagship you're told to be grateful you can access at all.

We're not anti-OpenAI or anti-Anthropic — we route work to whatever wins on cost and quality for the task. But a round this size is a signal worth reading: the open lane is now real infrastructure, not a hobbyist detour. Build so you can use it.

Key takeaways

  • Together AI raised $800M at an $8.3B valuation on July 1, led by Aramco's Prosperity7 Ventures
  • Annual bookings topped $1.15B as enterprises move workloads to open-weight models
  • Open models behind a standard API give you frontier-adjacent quality without flagship pricing or lock-in
  • Operator move: route routine work to the cheapest capable model, and keep it portable across hosts

Paying flagship prices for work an open model could do? We build automations that route each task to the cheapest model that clears the bar — and stay portable across hosts, so a price hike is a config change, not a rebuild. See our approach or estimate what you'd save.

Sources: BusinessWire, Pulse 2.0.

  • #together-ai
  • #open-source-ai
  • #inference
  • #ai-costs
  • #funding
TR

Tommy Rush — Founder, Rush Commerce

Operator turned builder. 15+ years running operations — now shipping the systems businesses run on. More

Get The Rush Report weekly — one email, zero fluff.