Average Ratings 0 Ratings
Average Ratings 0 Ratings
Description
Hy4 flash is a high-speed generative AI model designed to deliver advanced reasoning and multimodal understanding at a lower computational cost than larger frontier models. Developed as part of Tencent's HY4 family, the model combines hybrid attention, LatentMoE, and Engram architectural technologies to improve inference efficiency. Its architecture contains 224 billion parameters but activates approximately 12 billion parameters during processing, reducing the resources required for generating responses. A one-million-token context window enables users to analyze extensive documents, technical materials, software repositories, and other information-intensive inputs. Hy4 flash supports applications such as deep research, document processing, spreadsheet analysis, PowerPoint creation, frontend development, and visual information interpretation. Its agentic capabilities allow AI assistants to interact with tools, perform computer-based activities, and complete multistep productivity workflows. Tencent uses reinforcement learning and inference optimization techniques to shorten generated outputs and achieve decoding speeds above 150 tokens per second. The model is accessible through Tencent Cloud TokenHub, WorkBuddy, Yuanbao, WeCom, QQ, ima, and other supported Tencent services. By combining multimodal reasoning, efficient architecture, and flexible application integration, Hy4 flash provides developers and enterprises with a cost-effective foundation for AI-powered productivity and automation.
Description
Step 3.5 Flash is a cutting-edge open-source foundational language model designed for advanced reasoning and agent-like capabilities, optimized for efficiency; it utilizes a sparse Mixture of Experts (MoE) architecture that activates only approximately 11 billion of its nearly 196 billion parameters per token, ensuring high-density intelligence and quick responsiveness. The model features a 3-way Multi-Token Prediction (MTP-3) mechanism that allows it to generate hundreds of tokens per second, facilitating complex multi-step reasoning and task execution while efficiently managing long contexts through a hybrid sliding window attention method that minimizes computational demands across extensive datasets or codebases. Its performance on reasoning, coding, and agentic tasks is formidable, often matching or surpassing that of much larger proprietary models, and it incorporates a scalable reinforcement learning system that enables continuous self-enhancement. Moreover, this innovative approach positions Step 3.5 Flash as a significant player in the field of AI language models, showcasing its potential to revolutionize various applications.
API Access
Has API
Yes
API Access
Has API
Yes
Integrations
GitHub
No
Hugging Face
No
ModelScope
No
OpenRouter
Yes
Tencent Cloud
Yes
Tencent Hy
Yes
arXiv
No
Integrations
GitHub
Yes
Hugging Face
Yes
ModelScope
Yes
OpenRouter
No
Tencent Cloud
No
Tencent Hy
No
arXiv
Yes
Pricing Details
$0.070 per 1M tokens (input)
$0.070 per 1M tokens (input)
$0.281 per 1M tokens (output)
$0.281 per 1M tokens (output)
Free Trial
No
Free Version
No
Pricing Details
Free
Free Trial
No
Free Version
Yes
Deployment
Web-Based
Yes
On-Premises
No
iPhone App
No
iPad App
No
Android App
No
Windows
No
Mac
No
Linux
No
Chromebook
No
Deployment
Web-Based
Yes
On-Premises
No
iPhone App
Yes
iPad App
Yes
Android App
Yes
Windows
No
Mac
No
Linux
No
Chromebook
No
Customer Support
Business Hours
No
Live Rep (24/7)
No
Online Support
Yes
Customer Support
Business Hours
No
Live Rep (24/7)
No
Online Support
Yes
Types of Training
Training Docs
Yes
Webinars
No
Live Training (Online)
No
In Person
No
Types of Training
Training Docs
Yes
Webinars
No
Live Training (Online)
No
In Person
No
Vendor Details
Company Name
Tencent
Founded
1998
Country
China
Website
hy.tencent.ai/research/hy4-flash
Vendor Details
Company Name
StepFun
Founded
2023
Country
China
Website
static.stepfun.com/blog/step-3.5-flash/