DeepInfra
deepinfra.com
New
Low-cost serverless inference API for open-source LLMs, vision, and embeddings
- Free tier
- No
- Per month
- Usage
- MCP
- No
Why it's here
Low-cost serverless open-model inference with familiar API features; the listed facts show no free tier or MCP server.
Visit DeepInfra(opens in a new tab)
Last verified
LLM APIs
#16 of 20Low-cost serverless open-model inference with familiar API features; the listed facts show no free tier or MCP server.
| per month | Pay as you goUsageFrom $0.02 to $2.85 per 1M tokens across Flex/Standard tiers | EnterpriseCustomDedicated GPU instances, custom models, SLAs |
|---|---|---|
| Commercial use | ||
| Tool calling | ||
| OpenAI-compatible | ||
| No data training | ||
| Prod rate limits | ||
| JSON schema | ||
| Prompt caching | ||
| Image input | ||
| Context window | 1000000 tokens | 1000000 tokens |