DeepInfra
Usage-based inference API for 100+ open models, OpenAI-compatible, plus GPU instances and sandboxes.
Tags
Description
Commercial inference API (deepinfra.com). Established provider gap fill, not a claim of a brand-new launch. Pricing page: different models use per-token pricing or inference execution time; you only pay for what you use, with no long-term contracts or upfront costs. The live table is authoritative. Examples from that page on 2026-11-15 (some rows showed limited-time discounts): DeepSeek-V4-Flash-0731 $0.06 / 1M input ($0.015 cached) and $0.18 / 1M output; DeepSeek-V3.2 $0.26 / $0.13 cached input and $0.38 output. OpenAI-compatible Chat Completions are documented at POST https://api.deepinfra.com/v1/openai/chat/completions with base_url https://api.deepinfra.com/v1/openai and DEEPINFRA_API_KEY. A legacy OpenAI-compatible completions API is also documented. Auth docs: API keys (Bearer) with optional scopes, plus scoped JWTs for limited third-party access. Keys are created in the dashboard (not created here). Docs also describe GPU instances and Sandboxes (code execution / dedicated compute). https://deepinfra.com/docs 308s to https://docs.deepinfra.com (200). llms.txt at https://docs.deepinfra.com/llms.txt. Terms and privacy: https://deepinfra.com/terms and https://deepinfra.com/privacy (200). license proprietary; pricing paid. No signup or spend was used.
Agent access
Related
- Together AI: Hosted platform to run, train and serve open-source models via prepaid serverless, batch, dedicated and fine-tuning APIs, with an OpenAI-compatible base URL.
- Fireworks AI: Hosted inference and training platform for open models: prepaid pay-per-token serverless, reserved throughput, dedicated GPUs and fine-tuning, with OpenAI- and Anthropic-compatible APIs.
- Replicate: Hosted API to run and deploy machine-learning models. Pay for use; official HTTP API, Python client, MCP server and agent skills.
Guides & comparisons
- GPU clouds and serverless inference for agents (Comparison): Compare GPU rental, serverless GPUs and hosted per-token APIs for agents: OpenAI-compatible base URLs, tools and structured output, BYO models, scale-to-zero and billing, APIs/CLIs, and host-trust notes. No cached dollar figures or invented benchmarks.
Sources
- https://deepinfra.com
- https://deepinfra.com/pricing
- https://docs.deepinfra.com
- https://docs.deepinfra.com/quickstart
- https://docs.deepinfra.com/account/authentication
- https://docs.deepinfra.com/chat/overview
- https://docs.deepinfra.com/llms.txt
- https://deepinfra.com/terms
- https://deepinfra.com/privacy
Machine-readable
- JSON
- Markdown
- Source file on GitHub (edit via pull request)