Quick start

One endpoint. Familiar tools.

Use the OpenAI-compatible format your application already understands.

Deployed — the endpoint below is live; current model availability is shown on the status page

Full API referenceEvery endpoint, limit, price, and error code in one page.

Quick start

The request shape, model names, and rates below are live. Point any OpenAI-compatible client at the base URL with your API key.

Base URL: https://b300.powerchampion.ai/v1
Choose a model: glm-5.2-fp8
Endpoint status: Deployed

Protected access

API keys are issued per customer with prepaid balance (nano-USD precision). Check your balance anytime with GET /dashboard/billing/subscription using your key; top up with redeem codes.

API key: «redacted:sk-…» — request yours by email

Service status: 0 of 10 models onlineView status page

Available endpoints

POST /v1/chat/completions: Chat, reasoning, coding (GLM 5.2 FP8, GLM 5.3 Flash Uncensored, Qwen3-VL 30B)
POST /v1/images/generations: Text-to-image (Flux Schnell, Chroma1 HD)
POST /v1/audio/transcriptions: Speech-to-text (Whisper Large v3)
POST /v1/audio/speech: Text-to-speech (IndexTTS2)
POST /v1/videos: Text- and image-to-video, asynchronous jobs (Wan 2.2 TI2V-5B)
POST /v1/embeddings: Embeddings for RAG (BGE-M3)
POST /v1/rerank: Document reranking (BGE Reranker v2-m3)
GET /v1/models: List available models
GET /dashboard/billing/subscription: Check prepaid balance
POST /v1/redeem: Redeem a top-up code

Quick start — real requests against the live endpoint; substitute your own API key.

Text chat completions with GLM 5.2 FP8.

curl https://b300.powerchampion.ai/v1/chat/completions \
  -H "Authorization: Bearer «redacted:sk-…»" \
  -H "Content-Type: application/json" \
  -d '{"model":"glm-5.2-fp8","messages":[{"role":"user","content":"Hello"}]}'