Qwen: Qwen3 Max

Text input Text output
Author's Description

Qwen3-Max is an updated release built on the Qwen3 series, offering major improvements in reasoning, instruction following, multilingual support, and long-tail knowledge coverage compared to the January 2025 version. It delivers higher accuracy in math, coding, logic, and science tasks, follows complex instructions in Chinese and English more reliably, reduces hallucinations, and produces higher-quality responses for open-ended Q&A, writing, and conversation. The model supports over 100 languages with stronger translation and commonsense reasoning, and is optimized for retrieval-augmented generation (RAG) and tool calling, though it does not include a dedicated “thinking” mode.

Key Specifications
Cost
$$$$
Context
256K
Released
Sep 05, 2025
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Top P Seed Response Format Tool Choice Max Tokens Tools Presence Penalty Temperature
Features

This model supports the following features:

Response Format Tools
Performance Summary

Qwen3 Max, released by Qwen on September 5, 2025, demonstrates strong overall performance, particularly excelling in reliability. It consistently performs in the top tier for speed, ranking in the 65th percentile across six benchmarks, indicating efficient processing. Its pricing is moderate, positioned at the 40th percentile, offering a balanced cost-to-performance ratio. Notably, Qwen3 Max exhibits exceptional reliability with a 100% success rate across all benchmarks, meaning it consistently provides usable responses without technical failures. The model showcases remarkable accuracy in several key areas. It achieved perfect 100% accuracy in Email Classification, Ethics, and General Knowledge benchmarks, often being the most accurate model at its price point and speed. Its Reasoning capabilities are also very strong, scoring 90.9% accuracy. In Coding, it performed well with 90.6% accuracy, and it demonstrated solid Instruction Following at 66.7%. While no significant weaknesses were identified, its instruction following accuracy, though good, is not as high as its perfect scores in other categories. Overall, Qwen3 Max is a highly reliable and accurate model, particularly strong in knowledge, ethical reasoning, and classification tasks, making it well-suited for applications requiring high precision and dependable output.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $1.2
Completion $6
Input Cache Read $0.24

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
Alibaba
Alibaba | qwen/qwen3-max 256K $1.2 / 1M tokens $6 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Free Executions Accuracy Cost Duration
Other Models by qwen