Llama 4 Maverick

Llama 4 Maverick is Meta's high-performance open-source multimodal model, also using MoE architecture, with 400B total parameters and 17B active parameters. It excels in image understanding, coding, and reasoning, with performance comparable to GPT-4o. Maverick has an ultra-long context window of 10 million tokens, providing significant advantages in long-document analysis, codebase understanding, and multimodal dialogue.

免费开源

MMLU 87.2%, LMArena 综合排名与 GPT-4o 相当