Evaluation overview

Overall rank

#50

74.7 points · 180 models

Open-weight rank

#15

83.1 points · 109 models

Data coverage

35 items

1 source · updated through 2026-04-22

Domain performance

Coding#43 / 160

76.9

Research#56 / 174

83.9

Vision#22 / 65

83.4

Evaluation highlights

All evaluations

Claw-Eval Pass³

Coding · Qwen3.6-27B Technical Report

60.6%

1/ 8

CountBench

Multimodal · Qwen3.6-27B Technical Report

97.8%

1/ 8

MMLU-Redux

Reasoning · Qwen3.6-27B Technical Report

93.5%

18/ 102

C-Eval

Chinese · Qwen3.6-27B Technical Report

91.4%

14/ 64

SkillsBench Avg5

Coding · Qwen3.6-27B Technical Report

48.2%

1/ 8

MathVista mini

Multimodal · Qwen3.6-27B Technical Report

87.4%

3/ 15

Basic information

Context (input / output): 262K / 33K
Parameters: 27B
Release date: 2026-04
License: Apache 2.0
Input modalities: Text · Image
Output modalities: Text

Availability

vLLM · FP8Deployment options

FP8 · 33 GB VRAM

Official site HuggingFace

Qwen3.6-27BOpen

Evaluation overview

Domain performance

Evaluation highlights

Basic information

Availability

Qwen3.6-27BOpen

Evaluation overview

Domain performance

Evaluation highlights

Basic information

Availability

Evaluation overview

Domain performance

Evaluation highlights

Basic information

Availability

Related links

Evaluation overview

Domain performance

Evaluation highlights

Basic information

Availability

Related links