3/ The results:
- LlamaParse Agentic and Agentic Plus: 179/179, zero Boos
- Best frontier VLMs (GPT-6 Luna, GPT-6.1 Terra): 99.4%
- Open VLMs: 61% to 93%
LlamaParse Agentic scores 179/179 on document parsing benchmark
AILlamaIndex says its LlamaParse Agentic and Agentic Plus modes scored 179 of 179 with zero Boos on its benchmark. The post reports the best frontier VLMs, GPT-6 Luna and GPT-6.1 Terra, reached 99.4%, while open VLMs ranged from 61% to 93%.
@llama_index
Part of a thread · earlier post