Gemini 3.8 Flash vs Qwen3.5-35B-A3B

Gemini 3.8 Flash leads on the Visual Intelligence Index by 14.9 points, and Gemini 3.8 Flash scores higher on 7 of the 7 benchmarks where the two differ.

Model

Compared with

GoogleGemini 3.8 Flash80.6%Rank #2 AlibabaQwen3.5-35B-A3B65.7%Rank #7

At a glance

Measure	Gemini 3.8 Flash	Qwen3.5-35B-A3B
Visual Intelligence Index	80.6%	65.7%
Rank	#2 of 9	#7 of 9
Cost per taskMean over the benchmarks, at list price.	$0.0474	—
Latency per taskEach benchmark's median, averaged.	25.5s	14.3s
Tokens per task	60,794	25,755
Input / 1M tokens	$0.75	—
Output / 1M tokens	$3.75	—
Context window	1M	256K
Availability	Proprietary	Open weights

How cost and latency are measured

Capabilities

Each capability index is a weighted mean of the benchmarks that contribute to the capability.

Capability	Gemini 3.8 Flash	Qwen3.5-35B-A3B	Lead
Perception	79.8%	65.3%	+14.5
Spatial Reasoning	82.5%	65.8%	+16.7
Temporal Reasoning	81.3%	65.5%	+15.8
Action / Event Understanding	78.2%	62.2%	+16.0
Causal Reasoning	81.6%	70.1%	+11.6
Information Retrieval	79.4%	63.6%	+15.8

Benchmarks

Benchmark	Gemini 3.8 Flash	Qwen3.5-35B-A3B	Lead
Video-MME v2	71.3%	42.5%	+28.8
LVBench	86.4%	64.4%	+22.0
Perception Test	85.5%	66.7%	+18.7
NExT-QA	88.0%	83.4%	+4.6
Q-Bench Video	69.3%	66.9%	+2.4
EgoSchema	80.8%	77.2%	+3.6
UCF101-AD	63.5%	41.3%	+22.2

Usability

Gemini 3.8 Flash

4.0/ 5

Reachable1.0: There is an endpoint you can call without hosting anything.
First-party API — supported
Third-party API — supported
Portable0.0: You can run it yourself, and are not tied to one vendor.
Open weights — not supported
Self-hostable — not supported
Multimodal input1.0: It takes the footage directly, rather than frames you extracted.
Video ingestion — supported
Audio — supported
Programmable1.0: Output you can parse, and tools it can call on its own.
Structured output — supported
Tool calling — supported
Operable1.0: Usable interactively and in bulk, not only one call at a time.
Streaming — supported
Batch API — supported

Qwen3.5-35B-A3B

4.5/ 5

Reachable1.0: There is an endpoint you can call without hosting anything.
First-party API — supported
Third-party API — supported
Portable1.0: You can run it yourself, and are not tied to one vendor.
Open weights — supported
Self-hostable — supported
Multimodal input1.0: It takes the footage directly, rather than frames you extracted.
Video ingestion — supported
Audio — supported
Programmable1.0: Output you can parse, and tools it can call on its own.
Structured output — supported
Tool calling — supported
Operable0.5: Usable interactively and in bulk, not only one call at a time.
Streaming — supported
Batch API — not supported