DigitalOcean — pricing & performance
Models served
16
of 908 tracked models
Mean output speed
29
tokens per second, averaged over
15 measured endpoints
15 measured endpoints
Mean latency
0.9s
time to first token, averaged over
15 measured endpoints
15 measured endpoints
Cheapest blended
$0.14
per 1M tokens at a 3:1
input:output blend
input:output blend
Output speed on DigitalOcean
Median output tokens per second for each model this host serves, derived from OpenRouter endpoint telemetry. Higher is better. We do not run these measurements ourselves.
12 of 16 models ▾
☰⚙⇩
+Add model from specific provider
The Known Good
16 models on DigitalOcean
Full leaderboard →
| Model | Creator | Input $/M | Output $/M | Blended | Speed | TTFT | Context |
|---|---|---|---|---|---|---|---|
DeepSeek: DeepSeek V3.2
|
DeepSeek | $0.42 | $1.36 | $0.66 | 15 | 1.0s | 164k |
|
DeepSeek: DeepSeek V4 Flash
|
DeepSeek | $0.11 | $0.22 | $0.14 | 11 | 1.0s | 262k |
|
DeepSeek: DeepSeek V4 Pro
|
DeepSeek | $1.39 | $2.78 | $1.74 | 6 | 2.7s | 262k |
Meta: Llama 4 Maverick
|
Meta | $0.25 | $0.87 | $0.41 | 5 | 0.7s | 128k |
MiniMax: MiniMax M2.5
|
MiniMax | $0.23 | $0.90 | $0.39 | 69 | 0.4s | 66k |
MoonshotAI: Kimi K2.5
|
Moonshot AI | $0.38 | $2.02 | $0.79 | 11 | 1.0s | 262k |
|
MoonshotAI: Kimi K2.6
|
Moonshot AI | $0.76 | $3.20 | $1.37 | 53 | 0.7s | 262k |
NVIDIA: Nemotron 3 Super
|
NVIDIA | $0.21 | $0.46 | $0.27 | 9 | 0.7s | 1.00M |
OpenAI: gpt-oss-120b
|
OpenAI | $0.070 | $0.49 | $0.17 | 62 | 0.5s | 128k |
Qwen: Qwen3 Coder 30B A3B Instruct
|
Alibaba | $0.45 | $1.70 | $0.76 | 68 | 0.5s | 262k |
|
Qwen: Qwen3.5 397B A17B
|
Alibaba | $0.39 | $2.45 | $0.90 | — | — | 131k |
| Mi Xiaomi: MiMo-V2.5 | Xiaomi | $0.10 | $0.28 | $0.15 | 23 | 1.1s | 262k |
| Mi Xiaomi: MiMo-V2.5-Pro | Xiaomi | $0.60 | $3.00 | $1.20 | 52 | 0.9s | 262k |
Z.ai: GLM 5
|
Z.ai | $0.75 | $2.40 | $1.16 | 8 | 0.9s | 64k |
|
Z.ai: GLM 5.1
|
Z.ai | $0.97 | $4.30 | $1.81 | 21 | 1.2s | 164k |
|
Z.ai: GLM 5.2
|
Z.ai | $1.05 | $4.40 | $1.89 | 23 | 0.9s | 262k |
The Known Good
Source. Prices, endpoint availability and throughput for DigitalOcean are ingested from OpenRouter;
model metadata is enriched from models.dev. A blank cell means the figure was not published for that endpoint —
it never means zero. Blended price is (3 × input + output) ÷ 4, the same 3:1 blend used everywhere on this
site; see methodology. Quality scores are not provider-specific:
a model scores the same wherever it is hosted, so quality lives on the
model pages.