Llama 4 Maverick

Llama 4 Maverick

Meta馃嚭馃嚫meta-llama/llama-4-maverick
Compare

Llama 4 Maverick is Meta's high-capability open-weight Mixture-of-Experts model, offering strong multimodal (text and image) understanding, reasoning, and coding at efficient inference cost. It targets general-purpose assistant and agentic use cases needing frontier-adjacent quality in an open model.

Released
Apr 5, 2025
Input
TextImage
Output
Text
Support
ReasoningTool callingStructured outputs

Performance

Median latency and throughput measured across recent requests.

Throughput122 tok/s
Latency1.69s

Best at

The categories where this model ranks highest.

Image understanding

#122 路 top 73%

Legal & Government

#243 路 top 62%

Medicine & Healthcare

#243 路 top 62%

Writing, Literature, & Language

#244 路 top 58%

DomainScoreRank
Software & IT Services1358#250
Writing, Literature, & Language1312#244
Life, Physical, & Social Science1342#255
Entertainment, Sports, & Media1295#245
Business, Management, & Finance1317#258
Mathematical1319#247
Legal & Government1341#243
Medicine & Healthcare1345#243
Multimodal
Image understanding1145#122
CategoryWin rateRank
3D40.2%#134
Data Visualization38.4%#136
UI Component40.8%#141
Code Categories35.8%#145
Game Development33.7%#147
Website34.4%#149

Latest news about Llama 4 Maverick

Recent articles and community posts, linked to their original sources.