Developer-first, from the first request.

Design-partner phase. The public api.schneeai.com endpoint is activated at general availability. The examples below use https://api.schneeai.com as a placeholder — design partners receive a partner-specific endpoint and API key from us at onboarding. Replace the hostname in any snippet with your partner endpoint.

Quickstart

Send your first request through the AI Gateway.

curl https://api.schneeai.com/v1/chat/completions \
  -H "Authorization: Bearer $SCHNEEAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4o-mini",
    "messages": [
      { "role": "user", "content": "Summarize this report." }
    ]
  }'

The Gateway authenticates, applies your routing and limit policies, calls the selected provider, records structured usage, and returns the response.

New to SchneeAI? The Tutorial walks through streaming, named prompts, error handling, and cost tracking — copy-runnable code in curl, Python, and TypeScript.

REST API

The AI Gateway exposes an OpenAI-compatible chat completions endpoint, plus endpoints for prompt resolution, usage queries, and control-plane operations.

Full request and response shapes will live in the API Reference once the public surface stabilizes — currently in preparation.

Authentication

Requests use a bearer token issued by SchneeAI. Tokens carry a service, tenant, and user identity — so limits, logs, and access controls always reflect the caller. Never expose tokens in client-side code; route AI calls through your backend.

Examples

• Chat endpoint — minimal request with model: "gpt-4o-mini" and a single user message. ("auto" — where the router picks per routing policy — is on the roadmap; see Failover design.)
• Named prompt — call the Gateway with a registered prompt name and variables.
• Budget-aware call — pass a feature tag so spend lands on the right budget.

SDKs (coming soon)

Go and TypeScript SDKs are on the roadmap. Today, integrate with any HTTP client — the REST API is the source of truth and SDKs will follow its shape.

Cost Calculator

Estimate per-request and monthly costs across major LLM providers with the token cost calculator. To scope the token count first, paste your prompt into the Token Counter & PII Scanner — it also flags any secrets or PII before the request leaves your boundary.

Developer tools

• Model Directory — every model the Gateway routes to, with effective pricing and capability filters.
• Prompt Template Library — production-ready prompts for summarization, RAG, code review, extraction, and PII masking. Copy and run across models.
• Cost Calculator — per-token price comparison plus an interactive build-vs-buy TCO widget.
• LLM Pricing Benchmark — indicative per-token prices across OpenAI, Anthropic, Google, Mistral, Cohere, DeepSeek, Meta, and xAI. Updated monthly.
• Token & PII Scanner — tokenize prompts client-side and detect PII patterns before the request leaves your boundary.
• Error Handling — HTTP status reference, SchneeAI-specific error codes, retry strategy, and idempotency for the chat completions endpoint.
• API Request Builder — interactive builder that emits ready-to-run curl, Python, TypeScript, and Go snippets from the same form.
• Service Status — live status of Gateway, Console, Vault, and upstream providers.

Stay in the loop

New endpoints, breaking changes, and release notes land on the changelog first. Email [email protected] with “Subscribe” in the subject and we’ll add you to the announcements list.