Back

Alibaba

Qwen3 14B

ReasoningOpen weights
Intelligence
6.4
Rank 613 of 744 measured models
Coding
13.8
Rank 254 of 298 measured models
Output speed
26.2 t/s
Rank 394 of 403 measured models
Price/1M
$0.15
Rank 128 of 659 measured models
Context window
33K
Rank 1012 of 1199 measured models

Qwen3 14B (Reasoning) scores 6.4 on the Artificial Analysis Intelligence Index, ranking 613th of 744 measured models. The closest models on this index are Gemini 1.5 Pro (May '24) (6.4), Llama 3.2 Instruct 90B (Vision) (6.4), and Qwen2.5 Turbo (6.4). At $0.15 per million tokens (3:1 input/output blended price), it is more expensive than Qwen2.5 Turbo ($0.088). With an output speed of 26.2 t/s, it is slower than Qwen2.5 Turbo (84.9 t/s).

Artificial Analysis Indices
Reasoning
Intelligence6.4
Rank 613 of 744 measured models
Coding13.8
Rank 254 of 298 measured models
Math55.7
Rank 130 of 270 measured models
Agentic0.9
Rank 166 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 & Accounting6.6
Rank 206 of 218 measured models
Strategy & Ops3.9
Rank 210 of 218 measured models
Legal7.5
Rank 207 of 218 measured models
Engineering11.6
Rank 197 of 217 measured models
Economics11.5
Rank 213 of 225 measured models
Standard Benchmarks
MMLU Pro77.4%
Rank 152 of 349 measured models
GPQA Diamond60.4%
Rank 402 of 619 measured models
HLE4.5%
Rank 521 of 664 measured models
LiveCodeBench52.3%
Rank 142 of 343 measured models
SciCode30.7%
Rank 209 of 231 measured models
MATH-50096.1%
Rank 34 of 202 measured models
AIME 202476.3%
Rank 29 of 194 measured models
AIME 202555.7%
Rank 130 of 270 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