Skip to main content

Documentation Index

Fetch the complete documentation index at: https://wisdom-docs.juheapi.com/llms.txt

Use this file to discover all available pages before exploring further.

WisGate is an AI inference API relay service that gives developers unified, OpenAI-style REST access to multiple AI models from various providers — text LLMs, image generators, embeddings, and more — through a single, consistent interface. Instead of wiring separate SDKs or bespoke endpoints for different AI providers, WisGate lets you call different models by changing model strings and a few parameters. Why is that useful? Many teams build products that need fallback models, capacity bursts, or cost-optimized model selection. A relay layer simplifies provider management, routing, and billing — shifting the work of juggling model endpoints from your product code to the relay service.

​
What can you do with WisGate?

​
What capabilities are available (text, images, embeddings, multimodal)?

WisGate exposes the same categories of AI capabilities you’d expect from provider APIs:
  • Text / Chat completions (chat assistants, summarization, Q&A).
  • Image generation (text→image models from various providers).
  • Embeddings (semantic search, clustering, RAG pipelines).
  • Multimodal requests (models that accept text + images).
  • Streaming responses (real-time streaming for chat completions).
Because WisGate routes to the chosen provider/model, the precise feature set depends on the model you choose. Use the per-model documentation in WisGate’s model catalog to confirm capabilities and limits.

​
Support workflow automation platforms

WisGate integrates with low-code/no-code automation platforms and workflow tools that teams use to stitch AI into business processes:
  • Zapier: WisGate actions/triggers let you generate AI responses inside Zaps and connect to thousands of apps (Slack, Gmail, Google Sheets, CRM systems). This is useful for non-engineering automation of reporting, routing, or simple chatbots.
  • n8n: Verified nodes let you use WisGate inside n8n workflows to connect AI calls with databases, CRMs, and message platforms.
  • Make (formerly Integromat), Pipedream, Activepieces: WisGate connectors exist for these platforms, enabling integration with Google Sheets, Slack, GitHub, and many more via prebuilt workflows.
These integrations allow product, marketing and ops teams to embed AI outputs into everyday workflows without writing a full backend.

​
What developer tooling integrations exist?

  • GitHub / CI workflows: WisGate can be used inside GitHub Actions for tasks such as code generation, test orchestration and automated PR comment generation.
  • IDE plugins / assistants: WisGate can be integrated as a provider option in code assistants for VS Code/JetBrains, enabling inline code completions and assistant features.
  • Observability / monitoring integrations: Platforms like Langfuse provide tracing/observability for applications that call external model providers; guides exist for integrating WisGate with observability tools to capture prompts, responses and costs.

​
How do I get started with WisGate?

Getting started with WisGate follows the familiar pattern used by most modern API platforms: create an account, obtain an API key/token, read the docs, and make a first request. The platform also publishes quick-start guides that show how to mimic common patterns (for example, an OpenAI-style chat API) so you can port existing integrations quickly.

​
Step 1 — Sign up and obtain credentials

  1. Create an account on WisGate’s site. Sign up and get your API key from the dashboard.
  2. Get your API key from the WisGate console. Navigate to the API token section in your personal center, create a new token, and get your access credential: sk-xxxxx.

Get started quickly

Follow our quickstart guide to make your first API call in minutes.

​
Step 2 — Read the docs and pick a model

WisGate exposes many models and provides quick examples for the most popular ones (GPT-style chat, image generation). The API reference shows model names, capabilities, and recommended request formats. Because different vendors implement slightly different parameter and prompt semantics, WisGate’s abstraction attempts to provide a normalized surface while still passing vendor-specific options where needed.

​
Step 3 — Make a simple request (example)

WisGate supports request formats very similar to the common OpenAI Chat API shape, so porting code is straightforward. Here’s an example for text models: 
curl --location --request POST 'https://api.wisgate.ai/v1/chat/completions' \
--header 'Authorization: Bearer {{api-key}}' \
--header 'Content-Type: application/json' \
--data-raw '{
  "model": "gpt-4",
  "messages": [
    {
      "role": "user",
      "content": "Hello!"
    }
  ],
  "stream": false
}'

View API reference

Explore the complete API documentation with examples and endpoints.

​
Conclusion

WisGate addresses a real pain point: the operational complexity of using multiple competing AI providers. By offering a single, OpenAI-compatible gateway to multiple models, WisGate accelerates experimentation, centralizes billing and key management, and lets product teams focus on delivering value instead of wiring SDKs.