Together AI to serve 1.8T tokens in a day via OpenRouter alone amid OSS boom
togethercompute · x · 2026-09-26
Together AI says its serverless APIs for open-weight models have grown so fast that it is now one of the largest originators of open-model tokens in the US, with steep week-over-week growth reflecting OSS adoption. Via OpenRouter alone it will serve 1.8T tokens in a single day; getting started costs just $5. The team plans to publish more details on the infrastructure soon.
More from Infra
- Fal in talks to raise at $15B valuation; Fireworks weighs $30B round — steph_palazzolo · 2026-09-26
- Crusoe abandons $1.25B plan to power AI data centers with Boom turbines — TechCrunch AI · 2026-09-26
- Hyperscaler 2026-27 capex outlook raised ~66%, adding roughly $750B in AI spend — rohanpaul_ai · 2026-09-26
- OpenRouter now routes 10T+ tokens daily: the multi-model distribution bet and the coming agent fraud wave — Latent Space · 2026-09-26
- DGX Spark Handbook Released for Local Inference Setup — Thionne_WTZ · 2026-09-26
- Open-source Jev router cuts LLM token costs ~9x by routing Claude Code and Codex traffic — airesearch12 · 2026-09-26