SHELFCRITTER AI Digest

COMPANY

Together AI

companytopic-noteneocloud

Overview

Together AI is a neocloud that rents NVIDIA GPU clusters and hosts open-weight models on managed inference — one of a small handful of “OSS-inference-as-a-service” specialists sitting between the hyperscalers and the frontier labs. In July 2026 the company closed an $800M Series C at an $8.3B post-money valuation, taking the tier into hyperscaler-adjacent capital range while hyperscaler price cuts and dropping cache-read prices at frontier labs compress the arbitrage the specialist neoclouds live in.

Timeline

  • 2026-07-05-AI-Digest — Together AI closes an $800M Series C at $8.3B post-money — a 2.5× step-up from the $3.3B Series B in February 2025 — led by Aramco Ventures (Saudi Aramco’s corporate VC arm, distinct from the PIF sovereign fund) with NVIDIA, Vista, and General Catalyst participating. Reports ~$1.15B annual bookings (not GAAP revenue) and 3× growth in open-model usage. Narrow read: an OSS-inference-as-a-service tier is capitalized as a real category — the $800M cheque and the Aramco-led composition put Together AI on the same rough scale as the specialist neoclouds Meta targeted with Meta Compute on 2026-07-03-AI-Digest. Structural read the digest carries: capital flows say the neocloud tier is real; hyperscaler price cuts say the margin window is narrowing — Meta Compute, the June AWS H100 price adjustments, and Anthropic/OpenAI‘s own dropping cache-read prices are all compressing the arbitrage OSS-inference specialists live in. Read the $1.15B booking rate against a compressing per-token margin, not against a static one; the shape to watch is whether Together AI converts a scale advantage into gross-margin durability, or whether the next raise happens against a compressed multiple.

Key Developments

  1. $800M Series C at $8.3B Post-Money — Neocloud Tier Capitalized to Hyperscaler-Adjacent Scale (July 2026): 2.5× step-up from the February 2025 Series B, Aramco Ventures leading with NVIDIA / Vista / General Catalyst participating; ~$1.15B annual bookings and 3× open-model usage growth. The tier that hyperscalers previously described as “renting spare capacity” is now capitalized like a real distribution channel — but the same-quarter Meta Compute launch and hyperscaler cache-read price cuts compress the margin window from above. The next raise multiple is the empirical test of whether scale converts to gross-margin durability.

See also: NVIDIA, Meta, Anthropic, OpenAI, MOC - Major Companies, MOC - AI Infrastructure.