WanAPIs (wanapis)

Developer-focused AI API gateway with an OpenAI-compatible endpoint. One API key gives access to GPT, Claude, Gemini, DeepSeek, image, video, and audio models behind a single unified surface — with a model marketplace (345+ models, 13+ vendors), usage logs, quota management, transparent metering, multi-channel routing, and failover.

APIs.yml: apis.yml

Type

• x-type: company
• submitted via: api-search/network#34

APIs

• WanAPIs Unified AI API — https://api.wanapis.com/v1 — OpenAI-compatible REST gateway aggregating GPT, Claude, Gemini, DeepSeek, plus image/video/audio model providers. Docs · Model Marketplace · Pricing

Tags

LLM Gateway, AI API Gateway, OpenAI Compatible, Model Marketplace, LLM, GPT, Claude, Gemini, DeepSeek, Image Generation, Video Generation, Audio, Multimodal, Routing, Failover

Common Properties

• Website
• Documentation
• Model Marketplace
• Pricing

Artifacts

WanAPIs does not publish a machine-readable OpenAPI. https://api.wanapis.com/openapi.json returns the admin dashboard HTML. The artifacts below were hand-authored by API Evangelist from https://wanapis.com/docs, https://wanapis.com/model, and https://wanapis.com/pricing. They model the OpenAI-compatible surface that WanAPIs actually exposes plus the WanAPIs-native Responses and async task endpoints.

OpenAPI

• openapi/wanapis-openapi.yml — OpenAPI 3.1 covering /chat/completions, /responses, /completions, /embeddings, /images/generations, /images/edits, /audio/speech, /audio/transcriptions, /audio/translations, /models, /models/{model}, /tasks/{task_id}. Bearer auth.

Naftiko Capabilities

One self-contained capability per resource family. Each binds WANAPIS_API_KEY, consumes the upstream WanAPIs HTTP surface, and exposes both REST (/v1/...) and MCP adapters.

• capabilities/wanapis-chat.yaml
• capabilities/wanapis-responses.yaml
• capabilities/wanapis-completions.yaml
• capabilities/wanapis-embeddings.yaml
• capabilities/wanapis-images.yaml
• capabilities/wanapis-audio.yaml
• capabilities/wanapis-models.yaml
• capabilities/wanapis-tasks.yaml

JSON Schema

• json-schema/wanapis-chatcompletionrequest-schema.json
• json-schema/wanapis-chatcompletionresponse-schema.json
• json-schema/wanapis-chatmessage-schema.json
• json-schema/wanapis-embeddingrequest-schema.json
• json-schema/wanapis-imagerequest-schema.json

JSON-LD

• json-ld/wanapis-context.jsonld — schema.org-aligned context for Model, Provider, Channel, ChatCompletion, Embedding, ImageGeneration, Task, Quota, etc.

Examples

• examples/wanapis-chat-completion-example.json — chat completion against claude-opus-4-7.
• examples/wanapis-embedding-example.json — multi-input embedding call.
• examples/wanapis-image-generation-example.json — DALL·E-3 image generation.

Governance

• rules/wanapis-rules.yml — Spectral ruleset (OAS 3.1) enforcing base URL, bearer auth, operationId casing, and OpenAI-compatible request shape conventions.
• vocabulary/wanapis-vocabulary.yml — domain vocabulary (Gateway, Provider, Model, Channel, Quota, Failover, Metering, Task, Webhook Callback, ...).

Commercial

• plans/wanapis-plans-pricing.yml — pay-as-you-go per-model token pricing, $0.20 welcome credit. reconciled: false — rates change per upstream vendor.
• rate-limits/wanapis-rate-limits.yml — 429/503 guidance and quota model. reconciled: false.
• finops/wanapis-finops.yml — FOCUS 1.3-aligned FinOps view with per-model token meters, cache-hit, image, audio, balance-topup, and request-count meters.

Notes

• WanAPIs is OpenAI-compatible: any official OpenAI SDK works by setting base_url=https://api.wanapis.com/v1 and api_key=<wanapis_key>.
• All inference endpoints use Authorization: Bearer <wanapis_api_key>.
• Long-running image/video/audio jobs may return a task_id; poll /tasks/{task_id} or configure a webhook callback.
• Per-model pricing is dynamic. Always reconcile against https://wanapis.com/pricing before billing decisions.

Timestamps

• Created: 2026-05-27
• Modified: 2026-05-27

Maintainers

• Kin Lane — kin@apievangelist.com