Hy3 API pricing and setup
Hy3 (hy3) is a Tencent model. It is available through the relay.fast API with one relay.fast key and pay-as-you-go billing.
Prices checked from the live catalog. "From" is the lowest price among the relay.fast sources that currently serve the model. The price you pay depends on the source your request uses and can change on any day. The pricing page always shows current values.
| Official input price | $0.132 per 1M tokens |
|---|---|
| Official output price | $0.528 per 1M tokens |
| relay.fast input from | $0.0198 per 1M tokens |
| relay.fast output from | $0.0792 per 1M tokens |
| Cache read from | $0.00495 per 1M tokens |
| Lower than the official price by | up to 85% |
| Context window | 256K tokens |
| Endpoints | /v1/chat/completions |
Example request
Keep the key in the RELAY_FAST_API_KEY environment variable.
curl https://relay.fast/v1/chat/completions \
-H "Authorization: Bearer $RELAY_FAST_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"hy3","messages":[{"role":"user","content":"Hello"}]}'Use Hy3 in coding tools
Setup guides: Claude Code, Codex, Cursor, Cline, Kilo Code, OpenCode, Aider. The API docs have one-command installers.
Frequently asked questions
How much does the Hy3 API cost?
On 2026-11-05, relay.fast charges from $0.0198 per 1M input tokens and $0.0792 per 1M output tokens for Hy3, against the official $0.132 and $0.528. The price depends on the source your request uses and can change; https://relay.fast/pricing/hy3 shows the current value.
What is the context window of Hy3?
Hy3 accepts up to 256K tokens of context in the relay.fast catalog. Output limits depend on the model and client settings.
Which API endpoint does Hy3 use?
Call /v1/chat/completions with a relay.fast API key. OpenAI-compatible SDKs use the base URL https://relay.fast/v1.
Do I need a Tencent account to use Hy3?
No. You use a relay.fast account and API key. relay.fast is an independent API gateway and is not affiliated with Tencent.