Compare models

Context, limits, capabilities, price and use cases side by side — up to 4 models at a time. The URL holds your selection, so you can share it.

Qwen3.6 PlusQwen3.7 Plus Add a modelChat with these 2

AttributeHide identical rows	Qwen3.6 Plusqwen/qwen3.6-plus	Qwen3.7 Plusqwen/qwen3.7-plus10% off
Specs
Context window	1M	1M
Max output	66K	64K
Input modalities	TextImageVideo	TextImageVideo
Output modalities	Text	Text
Capabilities
Streaming	Supported	Supported
Tool calling	Supported	Supported
Structured output (JSON)	Supported	Supported
Image input	Supported	Supported
Thinking mode	Supported	Supported
Price / 1M tokens
Input	$ 0.32>= 256K$ 1.25	$ 0.32$ 0.29>= 256K$ 0.96$ 0.86
Output	$ 1.88>= 256K$ 7.50	$ 1.25$ 1.13>= 256K$ 3.75$ 3.38
Cache read	$ 0.032>= 256K$ 0.12	$ 0.032$ 0.029>= 256K$ 0.096$ 0.086
Cache write	$ 0.40>= 256K$ 1.57	$ 0.40$ 0.36>= 256K$ 1.20$ 1.08
What it is good at
Best for	Long-context knowledge work	Enterprise knowledge work Multimodal coding agents
Published benchmarks
Terminal-Bench 2.0pass_ratesetups differ by: harness	61.6%(setup not fully disclosed)	70.3%Terminus(setup not fully disclosed)
SWE-Bench Proresolved_rate	56.6%(setup not fully disclosed)	57.6%(setup not fully disclosed)
SWE-Bench Verifiedresolved_rate	78.8%(setup not fully disclosed)	77.7%(setup not fully disclosed)
AndroidWorldsuccess_rate	—	81%(setup not fully disclosed)
OSWorld Verifiedsuccess_rate	62.5%(setup not fully disclosed)	73.3%(setup not fully disclosed)
MRCR v2 128Kscore	—	91.7%(setup not fully disclosed)
MMMU Proaccuracy	78.8%(setup not fully disclosed)	79%(setup not fully disclosed)
Humanity's Last Exam (with tools)accuracy	50.6%(setup not fully disclosed)	—
GPQA Diamondaccuracy	90.4%(setup not fully disclosed)	90.3%(setup not fully disclosed)
MCPMarkscore	48.2%(setup not fully disclosed)	58.7%(setup not fully disclosed)
VideoMMMUaccuracy	—	85.4%(setup not fully disclosed)