Process massive datasets with Qwen3 Max Preview, featuring an expansive 252K context window for long-document analysis. This model delivers cost-effective pricing at $1.56/1M input and $7.80/1M output tokens, native tool calling support. Access Qwen3 Max Preview via the AI/ML API API with up to 33K output tokens.
Tokens
Tokens
Tokens
Qwen3 Max Preview by AI/ML API costs $1.56 per 1M input tokens and $7.80 per 1M output tokens. Cached reads cost $0.31 per 1M tokens.