# Hy3 API inference

Canonical page: https://probseek.com/models/hy3/

A language model for coding, multi-step reasoning, and reliable tool-driven workflows. A model by Tencent, selected for the ProbSeek v1 catalog.

## Specifications

- Catalog model ID: `tencent/hy3`
- Model author: Tencent
- Source context window: 256K
- Source parameters: 295B total / 21B active
- Model license: Apache 2.0
- Service: Managed API inference
- Input / 1M tokens: Not announced
- Output / 1M tokens: Not announced
- [Upstream model card](https://huggingface.co/tencent/Hy3)

This catalog describes models selected for ProbSeek v1. Confirm account access, endpoint features, and pricing before sending requests, and review the Terms of Service and Privacy & Data Policy. Capabilities, context windows, parameters, and licenses describe upstream models. Enabled API features and serving limits depend on your account. Token prices have not been announced. GPU rentals are planned and are not available in v1. Review the Terms of Service, Privacy & Data Policy, and any account-specific data-processing terms before using the service. ProbSeek does not use API inputs or outputs to train models. Inference currently runs in the United States (US). European Union (EU) inference is planned and is not currently available. ProbSeek does not log prompt or response content in routine inference logs. Operational metadata logs are retained for 7 days after creation and then deleted. They contain request IDs, model identifiers, token usage, latency, and status codes, not prompt or response content.

## Upstream capabilities

- Reasoning: Work through multi-step questions and tasks that benefit from deeper context.
- Coding: Explore code generation, explanation, debugging, and programming assistance.
- Tool use: Build model-assisted workflows around functions defined by your application.
- Chat: Create conversational interfaces and instruction-following experiences.

## API request examples

Illustrative request only. Confirm the full endpoint, enabled model ID, supported fields, and response format before running it. Copying or replaying an example on this website does not call an API.

Set PROBSEEK_CHAT_ENDPOINT to your full chat-completions URL and PROBSEEK_API_KEY in your server environment. Install the Python requests package for the Python example.

```python
import os
import requests

response = requests.post(
    os.environ["PROBSEEK_CHAT_ENDPOINT"],
    headers={"Authorization":
        "Bearer " + os.environ["PROBSEEK_API_KEY"]},
    json={
        "model": "tencent/hy3",
        "stream": False,
        "max_tokens": 256,
        "messages": [{"role": "user",
            "content": "What could I build with AI?"}]
    },
    timeout=60
)
response.raise_for_status()
print(response.json())
```

```bash
curl "${PROBSEEK_CHAT_ENDPOINT:?Set your full HTTPS chat endpoint}" \
  --fail-with-body --max-time 60 \
  -H "Authorization: Bearer $PROBSEEK_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "tencent/hy3",
    "stream": false,
    "max_tokens": 256,
    "messages": [{
      "role": "user",
      "content": "What could I build with AI?"
    }]
  }'
```

[Integration guide](https://probseek.com/quickstart/) | [Compare models](https://probseek.com/models/)