Qwen3.5 2B
Alibaba · 2 thinking configurations
Intelligence6.9
Input / 1M$0.02
Output / 1M$0.10
Context262.1K
Tariff: Requesty · Source: models-dev · USD per million tokens
Thinking configuration
What does more thinking buy you?
Compare this model’s measured thinking configurations.
Scroll the chart ↔ · tap a point for details
Efficient frontier: best measured score at each cost2 measured configurations · 0 without both values
Selected: thinking thinking · Click a point to switch configuration
Qwen3.5-2B is a compact yet capable model from Alibaba's Qwen3.5 series. It features a 262K token context window, support for 201 languages, thinking/reasoning mode, and tool calling for agentic workflows. A strong choice for prototyping, fine-tuning, and efficient multilingual deployments.
Intelligence IndexArtificial Analysis composite index, not a percentage. Different versions of the index are not directly comparable.
6.9 pointsGPQA DiamondGraduate-level scientific questions. Measures specialist reasoning rather than web-search quality.
45.6 %Humanity’s Last ExamDifficult questions across many disciplines, compared under the same evaluation setup.
2.6 %Terminal-Bench 2.1Agent tasks in a terminal environment. Results also depend on the evaluation harness.
3.0 %τ²-BenchTool use and interaction in assistance scenarios. Does not replace an evaluation of your integration.
69.0 %IFBenchVerifiable instruction following. A proxy for precision rather than a benchmark of creative style.
31.5 %MMMU-ProMultimodal understanding of academic problems that include images.
43.0 %Omniscience accuracyFactual accuracy in Artificial Analysis's Omniscience evaluation.
8.1 %Specifications from the catalog
Inputtext, image
Outputtext
Tool callingYes
Structured outputYes
SpeedNot available
First tokenNot available
Maximum output262.1K
Catalog listing date2026-03-10