Qwen3.5-2B

Name: Qwen3.5-2B
Author: Alibaba

Multimodal

Alibaba

Qwen3.5-2B is a 2 billion parameter vision-language model using Gated DeltaNet hybrid architecture with a 3:1 ratio of linear attention to full softmax attention. It supports 262K native context length and features both thinking and non-thinking modes.

Key Specifications

Parameters

2.0B

Context

Release Date

March 2, 2026

Average Score

51.9%

API Documentation Repository Model Weights

Timeline

Key dates in the model's history

Announcement

March 2, 2026

Last Update

September 12, 2026

Today

October 6, 2026

Technical Specifications

Parameters

2.0B

Training Tokens

Knowledge Cutoff

Family

Capabilities

MultimodalZeroEval

Benchmark Results

Model performance metrics across various tests and benchmarks

Reasoning

Logical reasoning and analysis

GPQA

Thinking mode • Self-reported

51.6%

Other Tests

Specialized benchmarks

AA-LCR

Thinking mode • Self-reported

25.6%

BFCL-V4

Thinking mode • Self-reported

43.6%

C-Eval

Thinking mode • Self-reported

73.2%

Global PIQA

Thinking mode • Self-reported

69.3%

IFBench

Thinking mode • Self-reported

41.3%

IFEval

Thinking mode • Self-reported

78.6%

Include

Thinking mode • Self-reported

55.4%

LongBench v2

Thinking mode • Self-reported

38.7%

MAXIFE

Thinking mode • Self-reported

60.6%

MMLU-Pro

Thinking mode • Self-reported

66.5%

MMLU-ProX

Thinking mode • Self-reported

52.3%

MMLU-Redux

Thinking mode • Self-reported

79.6%

MMMLU

Thinking mode • Self-reported

63.1%

Multi-Challenge

Thinking mode • Self-reported

33.7%

NOVA-63

Thinking mode • Self-reported

46.4%

PolyMATH

Thinking mode • Self-reported

26.1%

SuperGPQA

Thinking mode • Self-reported

37.5%

t2-bench

Thinking mode • Self-reported

48.8%

WMT24++

Thinking mode • Self-reported

45.8%

License & Metadata

License

apache_2_0

Announcement Date

March 2, 2026

Last Updated

September 12, 2026

Similar Models

All Models

Qwen2.5-Omni-7B

Alibaba

MM7.0B

Best score:0.8 (HumanEval)

Released:Mar 2025

Qwen2.5 VL 7B Instruct

Alibaba

MM8.3B

Released:Jan 2025

Qwen3 VL 4B Instruct

Alibaba

MM4.0B

Best score:0.8 (MMLU)

Released:Sep 2025

Price:$0.10/1M tokens

Qwen3.5-4B

Alibaba

MM4.0B

Best score:0.8 (GPQA)

Released:Mar 2026

Qwen3 VL 4B Thinking

Alibaba

MM4.0B

Best score:0.8 (MMLU)

Released:Sep 2025

Price:$0.10/1M tokens

North Micro Vision Instruct

Cohere

MM2.4B

Best score:0.5 (MMLU)

Released:Aug 2026

Gemma 3n E2B

Google

MM8.0B

Best score:0.5 (ARC)

Released:Jun 2025

Gemma 3 4B

Google

MM4.0B

Best score:0.7 (HumanEval)

Released:Mar 2025

Price:$0.02/1M tokens

Recommendations are based on similarity of characteristics: developer organization, multimodality, parameter size, and benchmark performance. Choose a model to compare or go to the full catalog to browse all available AI models.