DigitalOcean

Serverless inference on DigitalOcean's own GPU fleet

DigitalOcean, LLC is a New York cloud provider whose Gradient AI Platform serves open-weight models on its own H100 fleet rather than brokering them to someone else's. On Opper the route carries the DeepSeek V4 line, GLM 5 through 5.3, Kimi K2.5 to K3, Qwen3.5 and Qwen3.8-Max, NVIDIA's Nemotron 3 family, Gemma 4, MiMo, MiniMax, Ministral 3 and both gpt-oss sizes. US-hosted, with no region pinning on the serverless endpoint. A fit when you want a broad open-weight catalogue from a single well-known cloud vendor and US hosting is acceptable.

1 route26 modelsUS🇺🇸 HQ United States
digitalocean.com

Models on DigitalOcean

Every model we route through DigitalOcean. Compare residency, ZDR, training posture, and price at a glance — full data-handling detail per route below.

ModelRegionZero data retentionTrainingContextInputOutput
USNot by defaultNo1M$2.85$14.25
USNot by defaultNo1M$1.40$4.40
USNot by defaultNo1M$2.00$6.00
USNot by defaultNo1M$0.15$0.50
USNot by defaultNo1M$0.87$1.74
USNot by defaultNo1M$1.32$3.96
USNot by defaultNo262K$0.70$2.20
USNot by defaultNo1M$0.08$0.25
USNot by defaultNo1M$0.07$0.17
USNot by defaultNo262K$0.95$4.00
USNot by defaultNo262K$0.40$1.50
USNot by defaultNo164K$1.30$4.30
USNot by defaultNo64K$1.00$3.20
USNot by defaultNo262K$0.50$2.70
USNot by defaultNo66K$0.30$1.20
USNot by defaultNo131K$0.55$3.50
USNot by defaultNo164K$0.25$0.80
USNot by defaultNo128K$0.06$0.39
USNot by defaultNo128K$0.05$0.45
USNot by defaultNo256K$0.18$0.50
USNot by defaultNo128K$0.20$0.70
USNot by defaultNo262K$0.20$0.20
USNot by defaultNo1M$0.30$0.65
USNot by defaultNo131K$0.90$1.70
USNot by defaultNo66K$0.50$0.90
USNot by defaultNo128K$0.20$0.60

Data handling per route

DigitalOcean hosts on 1 route. Each route has its own privacy posture, residency, and GDPR terms. Postures are maintained by Opper with a last-verification timestamp.

United States🇺🇸

Zero data retention is not on by default — a moderation hold applies. No training on customer data. US; dpf sccs; DPA available.

Zero data retention
Not by default — a moderation hold applies. Derived from the logging and moderation facts.
Training
No training on customer data.
Logging
None
Moderation
Holds content for review
Caching
Content cached for replay
Subprocessor access
No subprocessor reads content
GDPR DPA
DPA available
Transfer mechanism
dpf sccs

Start building with 700+ models

One API key. Every major provider. Up and running in minutes.

Get startedView Documentation
DigitalOcean on Opper — ZDR, training posture, GDPR DPA, models | Opper AI