Hy3
Catalog modelA language model for coding, multi-step reasoning, and reliable tool-driven workflows. A model by Tencent, selected for the ProbSeek v1 catalog.
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. Access details · Privacy & Data Policy
Intelligence for your next application.
A language model for coding, multi-step reasoning, and reliable tool-driven workflows. Evaluate it on the tasks that matter to your product.
Explore its 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.
Know what you are building on
These capabilities describe the upstream model. Check your account documentation for confirmed availability, exact model IDs, enabled modalities, tools, and serving limits.
Read the source model card ↗A familiar place to start.
This example illustrates a chat-completions request. Set the full endpoint and API key provided with your account. Confirm the API contract before running it.
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())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?"
}]
}' Before your first request
- Install the Python requests package, or use cURL.
- Set PROBSEEK_CHAT_ENDPOINT to your full chat-completions URL.
- Keep PROBSEEK_API_KEY in your server environment.
- Confirm model ID, pricing, endpoint limits, and effective data terms; handle API errors.
Model specifications
Reference details from the upstream model documentation.
- Model
- Hy3
- Catalog model ID
- tencent/hy3
- Model author
- Tencent
- Source context window
- 256K
- Source parameters
- 295B total / 21B active
- Capabilities
- Reasoning, Coding, Tool use, Chat
- Model license
- Apache 2.0
- Service
- API inference
- Endpoint limits
- Not published