# Together AI — Introducing preemptible compute: the same compute, half the price

- Company: Together AI (together.ai)
- Announced: 2026-09-10
- Category: availability-change
- Coverage: not counted
- Announcement: yes
- Group: announcements
- Source: https://www.together.ai/blog/introducing-preemptible-compute-the-same-compute-half-the-price
- Record: https://forck.live/items/9953-introducing-preemptible-compute-the-same-compute-half-the-price
- Subject: Inference platform
- Pricing: billed sub-hourly at a flat 50% of the on-demand rate

Together AI announced the public preview of preemptible compute for its GPU Clusters, offering a 50% discount on on-demand rates for interruption-tolerant workloads. The preemptible nodes use the same NVIDIA accelerated compute, are billed sub-hourly, and provide a five-minute drain sequence before reclamation. The feature is available on Kubernetes clusters in all regions.

## Evidence

Verbatim from https://www.together.ai/blog/introducing-preemptible-compute-the-same-compute-half-the-price:

> Today we're announcing the public preview of preemptible compute for Together GPU Clusters, available on Kubernetes clusters in all regions.

---

Record: https://forck.live/items/9953-introducing-preemptible-compute-the-same-compute-half-the-price
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md