Qwen: Qwen3.5-9B
Qwen's mid-tier model. Context window: 262K tokens.
Scores by test
Methodology →What you need to know
Qwen3.5-9B is optimized for high-precision structured tasks, achieving perfect internal scores in tabular data processing, structured output, and persona consistency. Its 262K context window is fully leveraged, as evidenced by a 5/5 rating in long context performance, making it an efficient choice for analyzing extensive documents without losing coherence.
At a blended cost of $0.137 per million tokens, the model provides high-tier utility at a low price point. It outperforms its size class in multilingual capabilities and structured data handling, though it shows a relative weakness in simple classification tasks, where it scores only 3/5.
Use this model for applications requiring strict adherence to schemas, complex data extraction from tables, or long-form context processing on a budget. Skip this model if your primary use case is high-accuracy classification or if you require a model with superior strategic analysis capabilities.
Strengths — Top 3
Relative weaknesses — Bottom 3
Similar models