Hy3 preview

Tencent chatreasoningtool_use Long

API ID: tencent/hy3-preview-20260421

Input Price
$0.06
/1M tokens
Output Price
$0.21
/1M tokens
Cached Input
$0.02
Save 67%

About Hy3 preview

Hy3 preview is a budget-friendly general-purpose model from Tencent with ultra-long context (262k), suitable for conversations, content creation, and general AI tasks.

๐Ÿ’ฐ
Price Ranking
#663 lowest price among 1240 Chat models

Model Specifications

Context Length
262k
Max Output
โ€”
Release Date
2026-04-22
Capabilities
chat reasoning tool_use
Input Modalities
text
Output Modalities
text

Best For

  • Complex reasoning, math problems, multi-step logic
  • Conversations, content writing, general assistance

Consider Alternatives For

  • Image understanding (needs vision capability)
  • Simple Q&A (cheaper models available)

๐Ÿ’ฐ Real-World Cost Examples

Estimated monthly costs for common use cases

Personal AI Assistant
$0.09
/month
50 conversations/day, ~500 tokens each
Customer Service Bot
$2.83
/month
1000 tickets/day, ~800 tokens each
Data Analysis Pipeline
$3.94
/month
500 analyses/day, ~2k tokens each

Tencent Model Lineup

Compare all models from Tencent to find the best fit

Model Input Output Context Capabilities
Hy3 preview Current $0.06 $0.21 262k chat reasoning tool_use
Hunyuan A13B Instruct $0.14 $0.57 131k chat reasoning
Hy3 $0.14 $0.58 262k chat reasoning tool_use

Similar Models from Other Providers

Cross-brand alternatives with similar capabilities

Mistral Mistral Small 3.2 24B
Input: $0.07
Output: $0.20
Context: 256k
Mistral Devstral 2 2512
Input: $0.05
Output: $0.22
Context: 262k
Alibaba Qwen Qwen2.5 VL 32B Instruct
Input: $0.05
Output: $0.22
Context: 128k
Zhipu Ai GLM 4.5 Air
Input: $0.05
Output: $0.22
Context: 131k

๐Ÿš€ Quick Start

Get started with Hy3 preview API

OpenAI-compatible SDK
from openai import OpenAI

client = OpenAI(
    base_url="https://api.provider.com/v1",
    api_key="YOUR_API_KEY"
)

response = client.chat.completions.create(
    model="tencent/hy3-preview-20260421",
    messages=[
        {"role": "user", "content": "Hello!"}
    ]
)
print(response.choices[0].message.content)