Back

Alibaba

Qwen3.5 9B

ReasoningOpen weightsModels.devHugging Face
Intelligence
11.2
Rank 395 of 744 measured models
Coding
28.7
Rank 187 of 298 measured models
Output speed
69.5 t/s
Rank 254 of 403 measured models
Price/1M
$0.151
Rank 140 of 659 measured models
Context window
262K
Rank 458 of 1199 measured models
Cost per task
$0.213
Rank 73 of 193 measured models

Qwen3.5 9B (Reasoning) scores 11.2 on the Artificial Analysis Intelligence Index, ranking 395th of 744 measured models. The closest models on this index are Qwen3 Next 80B A3B (11.2), Qwen3 Next 80B A3B Thinking (11.2), and GLM-4.5-Air (11.1). At $0.151 per million tokens (3:1 input/output blended price), it is cheaper than Qwen3 Next 80B A3B ($0.412), Qwen3 Next 80B A3B Thinking ($0.413), and GLM-4.5-Air ($0.372). With an output speed of 69.5 t/s, it is about as fast as GLM-4.5-Air (71.1 t/s), slower than Qwen3 Next 80B A3B (163 t/s).

Artificial Analysis Indices
Reasoning
Intelligence11.2
Rank 395 of 744 measured models
Coding28.7
Rank 187 of 298 measured models
Agentic1.2
Rank 150 of 189 measured models
Reasoning levels
Each point is a reasoning level of this model: more reasoning often raises the score but costs more. Select a point to open that level.
Intelligence Indexmost efficient ↗Price / 1M tokens
Capability indexes
Performance in specific industries, measured by Artificial Analysis
Finance & Accounting9.8
Rank 188 of 218 measured models
Strategy & Ops6.6
Rank 189 of 218 measured models
Legal11.0
Rank 192 of 218 measured models
Engineering14.1
Rank 188 of 217 measured models
Economics15.5
Rank 198 of 225 measured models
Standard Benchmarks
GPQA Diamond80.6%
Rank 202 of 619 measured models
HLE14.9%
Rank 260 of 664 measured models
SciCode29.5%
Rank 211 of 231 measured models
IFBench66.7%
Rank 92 of 450 measured models
AA-LCR70.0%
Rank 218 of 575 measured models
TerminalBench Hard24.2%
Rank 151 of 432 measured models
Terminal-Bench 2.129.2%
Rank 163 of 244 measured models
Terminal-Bench v40.5%
Rank 136 of 218 measured models
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.

Preferences and analytics

Rybbit: visits, performance, clicks, copies, forms and session replay (masked inputs), with your permission. Refusing does not limit the site.

Privacy

Allow navigation analytics?

With your permission, Rybbit measures visits, performance and interactions (clicks, copies, forms), including session replay, to improve BenchSift. Input values are masked in replays. Refusing does not limit the site.

Your choice is stored for 180 days. Change it at any time using “Privacy preferences” in the footer. Withdrawal reloads the page.

No choice stored: Rybbit is disabled.

Privacy