Models
Every model. One key.
ModelServer gives you OpenAI-compatible access to 9+ frontier models from 4 providers. Pay only for what you use, with a flat 5.5% platform fee.
Anthropic
3 models
Claude Opus 4
200KAnthropic's flagship reasoning model. Best for complex analysis, long-form writing, and agentic tasks.
Input
$15.00
Output
$75.00
/ 1M tok
Claude Sonnet 4
200KBalanced quality and speed. The default Claude for production workloads.
Input
$3.00
Output
$15.00
/ 1M tok
Claude Haiku 4.5
200KFastest Claude — sub-second latency for high-throughput applications.
Input
$1.00
Output
$5.00
/ 1M tok
OpenAI
3 models
GPT-4o
128KOpenAI's flagship multimodal model — text, vision, and structured output.
Input
$2.50
Output
$10.00
/ 1M tok
GPT-4o mini
128KCheap and fast OpenAI workhorse. Great default for high-volume apps.
Input
$0.15
Output
$0.60
/ 1M tok
o3
200KOpenAI's reasoning model. Step-by-step problem solving for math, code, and science.
Input
$10.00
Output
$40.00
/ 1M tok
1 models
DeepSeek
2 models