Hy3
Hy3 is a text generation model from Tencent with a 262,144-token context window, available via DeepInfra.
Tencent
DeepInfraLarge-context text generation from Tencent
Hy3 is a large language model developed by Tencent and released in July 2026. It is hosted on DeepInfra under the identifier tencent/Hy3 and is classified as an open-source model, meaning its weights are publicly available for inspection and use. The model supports a context window of 262,144 tokens and can generate responses up to 32,768 tokens in length.
Hy3 is suited for tasks that require processing long documents, extended conversations, or large codebases within a single context. Its open-source status makes it accessible for developers who want to self-host or fine-tune the model for specific applications. The model is tagged as low cost, which reflects its positioning for use cases where inference budget is a consideration.
What Hy3 supports
Large Context Window
Supports up to 262,144 tokens of context in a single request, enabling processing of long documents, books, or extended conversation histories.
Text Generation
Generates natural language responses for tasks such as summarization, question answering, and drafting, with a maximum response size of 32,768 tokens.
Open Source Weights
Model weights are publicly available, allowing developers to self-host, inspect, or fine-tune Hy3 for custom applications.
Low-Cost Inference
Positioned as a low-cost model on DeepInfra, making it suitable for high-volume or budget-sensitive inference workloads.
Ready to build with Hy3?
Get Started FreeCommon questions about Hy3
What is the context window size for Hy3?
Hy3 supports a context window of 262,144 tokens, allowing it to process very long inputs in a single request.
What is the maximum response length Hy3 can generate?
Hy3 can generate responses of up to 32,768 tokens per request.
Is Hy3 an open-source model?
Yes, Hy3 is tagged as open source, meaning its weights are publicly available for use, inspection, and fine-tuning.
Who developed Hy3 and when was it released?
Hy3 was developed by Tencent and released in July 2026. It is served via DeepInfra under the identifier tencent/Hy3.
Does Hy3 support image or video inputs?
Based on the available metadata, Hy3 does not have confirmed support for image or video inputs. It is classified as a text generation model.
Documentation & links
Parameters & options
Explore similar models
AmazonAmazon Nova 2 Lite
Amazon Nova 2 Lite is a text and image-input model from Amazon with a 1,000,000-token context window.
GoogleGemini 3.7 Flash
Gemini 3.7 Flash is a text generation model from Google with a 1,048,576-token context window and configurable thinking levels.
AnthropicClaude 5 Opus
Claude 5 Opus is a text generation model from Anthropic with a 1,000,000-token context window and native tool use.
MetaMuse Spark 1.1
Muse Spark 1.1 is a text generation model from Meta supporting images and a 1,048,576-token context window.
Z.aiGLM-5.2
GLM-5.2 is a text generation model from Z.ai with a 1,048,576-token context window and up to 128,000 token responses.
MiniMaxMiniMax M3
MiniMax M3 is a multimodal text generation model from MiniMax with a 524,288-token context window.
Start building with Hy3
No API keys required. Create AI-powered workflows with Hy3 in minutes — free.