Docs
Tencent Hy4 API Guide: Endpoints & cURL
Tencent Hy4 is currently "In preview". This guide provides model IDs, price snapshots, and copyable examples for integration. Tencent Hy4 model page
Overview
- Vendor: Tencent
- Type: Text
- Number of tiers: 1
- Endpoint tags: openai, openai-response, anthropic
Available tiers and pricing
The model in requests must use the model IDs in the table below, case-sensitive.
hy4 · 1,024,000 tokens
Catalog reference rates · billed on actual token usage
- Input tokens
- $0.50 / 1M tokens
- Output tokens
- $1.50 / 1M tokens
More price combinations
- Cached input
- $0.50 / 1M tokens
- Cache write
- $0.625 / 1M tokens
- Cache write (1 hour)
- $1.00 / 1M tokens
- Image input tokens
- $0.50 / 1M tokens
- Audio input tokens
- $0.50 / 1M tokens
- Audio output tokens
- $0.50 / 1M tokens
Context tiers are not listed. Reference rates may not apply at every context length.
Snapshot time Final charges depend on the actual request and settlement.
Base URL and authentication
Base URL is https://api.rokoapi.com/v1. All requests must include Authorization: Bearer $OPENAI_API_KEY.
Quickstart
curl https://api.rokoapi.com/v1/chat/completions \
-H "Authorization: Bearer $OPENAI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "hy4",
"stream": true,
"messages": [{"role": "user", "content": "Introduce yourself in one sentence"}]
}' Common parameters
| Parameter | Description |
|---|---|
model | Use the public model IDs from the table above. |
prompt / messages | Input prompt, conversation messages, or generation instructions; format depends on the endpoint. |
stream | Text models can set this to true to enable SSE streaming output. |
Notes
- Use only the model IDs listed on this page; unprovisioned or retired names will return a model unavailable error.
- If you encounter a 403 error, check your account balance, key group, and model access permissions.
- Multimodal, tool calling, caching, and streaming capabilities depend on the upstream model and channel configuration; refer to actual responses.