EtaYang10th/Open-M3-Bench

M3-Bench Image Assets This dataset contains the image assets for M3-Bench: Multi-Modal, Multi-Hop, Multi-Threaded Tool-Using MLLM Agent Benchmark, a benchmark for multimodal tool use under the Model Context Protocol (MCP). M3-Bench targets realistic workflows where an agent must jointly reason over images, text, and MCP tools in multi-hop and sometimes multi-threaded trajectories. For full details, please see the paper: https://arxiv.org/abs/2511.17729. Contents… See the full description on the dataset page: https://huggingface.co/datasets/EtaYang10th/Open-M3-Bench.

<iframe src="https://silytics.si/embed/item/8baa1f25-7568-419f-917c-f71c16cf2924" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.