{"url":"/dataset/toolbench","name":"ToolBench","full_name":null,"description_markdown":"**ToolBench** is an instruction-tuning dataset for tool use, which is created automatically using ChatGPT. Specifically, the authors collect 16,464 real-world RESTful APIs spanning 49 categories from RapidAPI Hub, then prompt ChatgPT to generate diverse human instructions involving these APIs, covering both single-tool and multi-tool scenarios.","description_withheld":null,"homepage":"https://github.com/OpenBMB/ToolBench","introduced_date":"2023-07-31","introduced_date_note":null,"introduced_by":{"paper":"/paper/toolllm-facilitating-large-language-models-to","title":"ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs","first_author":"Yujia Qin","url":null},"license":{"name":"Apache-2.0 license","url":"https://github.com/OpenBMB/ToolBench/blob/master/LICENSE"},"modalities":[{"name":"Texts","url":"/datasets/modality/texts"}],"tasks":[{"name":"Trajectory Planning","url":"/task/trajectory-planning","datasets_with_task":"/datasets/task/trajectory-planning"}],"languages":[{"name":"English","url":"/datasets/language/english"}],"variants":["ToolBench"],"data_loaders":[],"num_papers_in_archive":105,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/trajectory-planning-on-toolbench","task":"Trajectory Planning","dataset_variant":"ToolBench","rows":3,"metrics":["Win rate"],"first_row_in_archive_order":{"model":"GPT4-TOPGUN","paper":"/paper/swissnyf-tool-grounded-llm-agents-for-black","metrics":{"Win rate":"86.54"},"code_links":[{"title":"iclr-dummy-user/swissnyf","url":"https://github.com/iclr-dummy-user/swissnyf"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/swissnyf-tool-grounded-llm-agents-for-black","title":"SwissNYF: Tool Grounded LLM Agents for Black Box Setting","date":"2024-02-15","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":2,"samples_ran":2,"samples_unverified":0,"pointer_only_for_licence":2,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/fortify-the-shortest-stave-in-attention","title":"Fortify the Shortest Stave in Attention: Enhancing Context Awareness of Large Language Models for Effective Tool Use","date":"2023-12-07","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":4,"samples_ran":4,"samples_unverified":0,"pointer_only_for_licence":4,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/toolllm-facilitating-large-language-models-to","title":"ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs","date":"2023-07-31","rows_on_this_dataset":1,"code_links":2,"syntology":null}],"syntology_totals":{"read_at":"2026-09-24T18:15:14+00:00","papers_with_samples":2,"samples_harvested":6,"samples_ran":6,"samples_unverified":0,"pointer_only_for_licence":6,"papers_with_no_sample_that_ran":0,"note":"the per-paper counts above, summed; not a rate"},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks."}