Gemini 3 Flash vs Llama 4 Maverick
Google vs Meta — benchmarks, pricing, and capabilities side by side.
- •Gemini 3 Flash has the higher intelligence index (90.2 vs 63.9)
- •Llama 4 Maverick is cheaper ($0.15 vs $0.50 per 1M input)
- •Llama 4 Maverick is faster
| Gemini 3 Flash | Llama 4 Maverick | |
|---|---|---|
| Intelligence index | 90.2 | 63.9 |
| Developer | Meta | |
| Type | Multimodal | Multimodal |
| Access | API only | Open weights |
| Context window | 1,048,576 tokens | 1,048,576 tokens |
| Input price | $0.50 / 1M | $0.15 / 1M |
| Output price | $3.00 / 1M | $0.60 / 1M |
| Speed | 191 tok/s | 639 tok/s |
| Released | December 17, 2025 | April 5, 2025 |
| Parameters | — | 400B total / 17B active (MoE) |
| Input modalities | Text, Image, Audio, Video | Text, Image |
| Output modalities | Text | Text |
Shared benchmarks
Gemini 3 Flash
Llama 4 Maverick
AIME 2025
97
19.3
GPQA Diamond
90.4
69.8
Humanity’s Last Exam
34.7
4.8
LiveCodeBench
90.8
43.4
MMLU-Pro
89
80.5
SWE-bench Verified
78
30