ByteDance: Doubao-pro-32k API
The flagship Pro-tier model from ByteDance, optimized for complex reasoning, role-playing, and high-concurrency enterprise tasks. It features a sparse MoE architecture and a "Deep Thinking" mode, offering GPT-4o class performance at a fraction of the cost within a 32k window.
- Context window: 32,768 tokens
- Max output: 4,096 tokens
- Input: text, image, audio
- Output: text
- Reasoning: Supported
- Tool calling: Supported
- File input: Supported
- Released: 2025-01-21
- Knowledge cutoff: 2025-08-31
Frequently Asked Questions
What is the context window of Doubao-pro-32k?
Doubao-pro-32k supports a context window of up to 32,768 tokens.
Does Doubao-pro-32k support function calling?
Yes. Doubao-pro-32k supports tool / function calling.
Does Doubao-pro-32k support reasoning?
Yes. Doubao-pro-32k is a reasoning-capable model.
What is the knowledge cutoff of Doubao-pro-32k?
The knowledge cutoff of Doubao-pro-32k is 2025-08-31.