Llama 4 Maverick is Meta's high-capacity multimodal language model built on a Mixture-of-Experts architecture with 128 experts and 17 billion active parameters per forward pass (400B total). It supports multilingual text and image input across 12 languages with a 1-million-token context window, and uses early fusion for native multimodality. Instruction-tuned for assistant-like interaction and image reasoning, Maverick is optimized for vision-language tasks requiring advanced multimodal understanding and high throughput.
API|Vision|Open ModelLlama
Knowledge Cutoff
2024-08-31
Input → Output Format
Context Memory
1.0MIN16KOUT
AI Performance Evaluation
Arena Overall Score
1327
±4As of 2026-04-23
Overall Rank
No.185
40,010 Votes
Arena by Ability
Hard Prompts
1338±6No.180
Expert Knowledge
1324±13No.169
Instruction Following
1312±6No.182
Conversation Memory
1324±8No.178
Creative
1307±9No.164
Coding
1372±7No.174
Math
1319±11No.167
Arena by Occupation
Creative Writing
1311±7No.174
Social Sciences
1344±8No.182
Media
1295±8No.175
Business
1320±8No.184
Healthcare
1347±13No.173
Legal
1343±12No.175
Software
1359±6No.178
Mathematics
1320±11No.170
Source:Arena Intelligence
Overall
AA Intelligence Index
18%↓20%
ForecastBench
57%↓2%
Reasoning & Math
AA Math Index
19%↓54%
GPQA Diamond
67%↓14%
HLE
4.8%↓12%
MMLU-Pro
81%↓1%
AIME 2025
19%↓54%
MATH-500
89%↓4%
Coding
AA Coding Index
16%↓18%
LiveCodeBench
40%↓26%
TAU2
18%↓55%
TerminalBench
6.8%↓24%
SciCode
33%↓8%
Language & Instructions
IFBench
43%↓14%
AA-LCR
46%↓16%
Hallucination (HHEM)
8.2%↓2%
Factual (HHEM)
92%↑2%
Output Speed
Standard Mode
119tok/s↑37
First Output 0.67s