Back

Google

Gemma 3 4B Instruct

Open weightsHugging Face
Intelligence
4.8
Rank 725 of 744 measured models
Coding
2.7
Rank 294 of 298 measured models
Context window
128K
Rank 773 of 1199 measured models

Gemma 3 4B Instruct scores 4.8 on the Artificial Analysis Intelligence Index, ranking 725th of 744 measured models. The closest models on this index are Apertus 8B Instruct (4.8), Gemma 3 1B Instruct (4.8), and Granite 4.0 350M (4.8).

Artificial Analysis Indices
Intelligence4.8
Rank 725 of 744 measured models
Coding2.7
Rank 294 of 298 measured models
Math12.7
Rank 229 of 270 measured models
Standard Benchmarks
MMLU Pro41.7%
Rank 319 of 349 measured models
GPQA Diamond29.1%
Rank 586 of 619 measured models
HLE5.3%
Rank 455 of 664 measured models
LiveCodeBench11.2%
Rank 311 of 343 measured models
MATH-50076.6%
Rank 122 of 202 measured models
AIME 20246.3%
Rank 150 of 194 measured models
AIME 202512.7%
Rank 229 of 270 measured models
IFBench28.3%
Rank 396 of 450 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