Model decision surface

Compare AI models

Start with Qwen3.5-9B and Qwen3 VL 32B Instruct, or choose any two current ranked language models. Compare capability evidence, price, context, provider availability, and weight access without pretending one field decides every use case.

First modelSecond model

Current public data

Qwen3.5-9B vs Qwen3 VL 32B Instruct

Live dataset updated 10/11/2026, 3:39:45 PM UTC

Open 1200×630 evidence receipt

Factual comparison of Qwen3.5-9B and Qwen3 VL 32B Instruct
Field	Qwen3.5-9B Qwen	Qwen3 VL 32B Instruct Qwen
ModelCap position	#132	#154
Index score	45.9	41.7
Evidence	Estimatedpublisher-corpus-prior over 211 held-out anchors (31%); launch card against 3 resolved peers on 10 rows (69%); exceeds every named peer on 4 of 10 rows; optimism haircut 0.6 from cross-lab-probe; shrunk 2.6 up toward the measured corpus	Estimated4 reported rows against measured corpus ladders (51%); publisher-corpus-prior over 211 held-out anchors (50%); shrunk 9.8 up toward the measured corpus
Input / 1M	$0.10	Unavailable
Output / 1M	$0.15	Unavailable
Pricing status	fresh	empty
Context	262K	131K
Providers	6	0
Weight access	Open weights	Open weights

Decision facts

Qwen3.5-9B is #132; Qwen3 VL 32B Instruct is #154 on the same current language board.
Index scores are 45.9 for Qwen3.5-9B and 41.7 for Qwen3 VL 32B Instruct. Their published uncertainty intervals overlap, so the rank difference alone does not establish a reliable capability advantage for your workload.
Both positions use Estimated evidence.
Published context is 262,144 tokens for Qwen3.5-9B and 131,072 for Qwen3 VL 32B Instruct.
ModelCap currently lists 6 providers for Qwen3.5-9B and 0 for Qwen3 VL 32B Instruct.

These are separate published fields, not a synthetic winner. ModelCap does not collapse price, access, context, and capability evidence into a hidden recommendation score.

Popular comparisons

Each comparison page is a permanent, shareable URL with the same live figures as this tool: ModelCap Index position, API pricing, context window, provider count, weight access and every shared benchmark board.