Providers /DigitalOcean Gradient
Inference provider · HQ US

DigitalOcean Gradient

Models
14
Data residency
✓EU endpoint
00

Our take

Editorial by LLMap · updated Sep 5, 2026

DigitalOcean Gradient is a cloud inference service with strong compliance credentials and measured throughput on most of its 20-model catalogue. It suits teams that need verified data-sovereignty options and want to pick models by speed as well as price.

Use this for compliance-sensitive deployments that need SOC 2, a zero-retention option, no prompt training, or an EU-region endpoint. Pick it when throughput matters: 11 of 12 sampled offers have verified tokens-per-second data, and Nemotron 3 Super hits 41 tokens/second. Consider it for budget input on GPT-OSS 120B. Skip it if you need a very wide model catalogue, or if you rely on models with unverified speed data.

Strengths
  • Strong compliance and data-sovereignty posture — SOC 2 attestation, zero-retention option, no prompt training, and EU-region endpoint all verified.
  • Cheapest tracked input price for OpenAI GPT-OSS 120B.
  • Highest measured throughput on NVIDIA Nemotron 3 Super at 41 tokens/second, 4.6× faster than DeepSeek V4 Flash and 10.3× faster than Qwen3 Coder 30B.
  • Broad throughput transparency: 11 of 12 sampled offers have verified tokens/second.
Trade-offs
  • Some models show very low throughput — Qwen3 Coder 30B at 4 tokens/second and Qwen3.5 397B at 6 tokens/second are the two slowest measured offers.
  • Output prices can spike well above input prices: GPT-OSS 120B output is 7.0× its input price, and Qwen3.5 397B output is 6.4×.
  • Smaller catalogue than major open-weight hosts — 20 tracked offers.
01

Point your tools here

We have not recorded what a router needs for DigitalOcean Gradient yet — no base URL and no docs link on file. That is our gap, not a sign DigitalOcean Gradient has no API; its own documentation is the place to look until we close it.

02

Privacy & data handling

These answers cover requests sent to DigitalOcean Gradient through OpenRouter, as OpenRouter records them, last read 4 hours ago. For DigitalOcean Gradient's own API we hold no answer: check its terms. SOC 2 and the data processing agreement below are hand-checked, not part of that sync.

Trains on prompts
✗No
Logs prompts
✗No
✓Offered
✓Attested

We have not dated this attestation or linked the report yet.

No answer on record

Prompt retention: none.

03

Models & pricing

Something wrong on this page? Tell us