OpenAI compatible API · Attested · Public status

DigitalOcean Gradient AI performance

Measured TTFT, TTFB, effective throughput, uptime, and sampled model routes for DigitalOcean Gradient AI.

Verify gateway
Onebase URL to migrate
100sof models and routes
0prompt or output logs. Always.

digitalocean

65 samples

Provider overview

Continuously sampled provider performance. TrustedRouter reports unsupported route and probe-configuration rows separately from provider downtime. Prompt and output content is not stored.

p50 TTFT1572 ms
p95 TTFT5313 ms
p50 TTFB1689 ms
Effective throughput28 tok/s n=7
Uptime98.46%

Measured model routes

Modelp50 TTFTp50 TTFBEffective throughputUptimeConfig excludedAvailability samples
minimax/minimax-m2.5 651 ms 651 ms 100.00% 1
nvidia/nemotron-3-super-120b 898 ms 898 ms 100.00% 2
nvidia/nemotron-3-nano-omni 962 ms 961 ms 100.00% 4
qwen/qwen3.5-397b-a17b 965 ms 965 ms 100.00% 2
z-ai/glm-5 1341 ms 1341 ms 100.00% 3
qwen/qwen3-coder-flash 1399 ms 1399 ms 100.00% 6
meta-llama/llama-4-maverick 1429 ms 1429 ms 100.00% 5
moonshotai/kimi-k2.6 1474 ms 1474 ms 29 tok/s n=1 100.00% 6
z-ai/glm-5.1 1554 ms 1553 ms 100.00% 3
moonshotai/kimi-k2.5 1572 ms 1572 ms 100.00% 4
nvidia/nemotron-nano-12b-v2-vl 1586 ms 1586 ms 100.00% 6
deepseek/deepseek-v3.2 1596 ms 1596 ms 100.00% 3
nvidia/nemotron-3-ultra-550b 1637 ms 1637 ms 100.00% 3
meta-llama/llama-3.3-70b-instruct 1689 ms 1689 ms 100.00% 2
deepseek/deepseek-r1-distill-llama-70b 3010 ms 3010 ms 100.00% 2
mistralai/ministral-3-14b-instruct 3109 ms 3109 ms 100.00% 1
deepseek/deepseek-v4-flash 3423 ms 3423 ms 20 tok/s n=1 100.00% 3
deepseek/deepseek-v4-pro 3489 ms 3489 ms 26 tok/s n=2 100.00% 6
google/gemma-4-31b-it 4371 ms 4371 ms 100.00% 1
z-ai/glm-5.2 5351 ms 5351 ms 28 tok/s n=2 50.00% 2
xiaomi/mimo-v2.5-pro 64 tok/s n=1 0
Workspace access

Sign in

Choose a sign in method. New email and OAuth accounts include $0.10 in starter credit; wallet-only accounts start at $0.

By signing in you agree to the terms of service and privacy policy.