Datasets:

sauravsingla08
/

AgentWeave-Tool-Routing

id int64 1 40	query stringlengths 34 68	dataset stringclasses 1 value	description stringclasses 1 value
1	research and summarize evidence about a technical topic	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
2	analyze and verify a technical claim using multiple sources	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
3	plan and optimize an engineering workflow	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
4	retrieve research then analyze and summarize the findings	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
5	code analyze and verify a solution	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
6	research plan analyze and summarize a complex requirement	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
7	retrieve evidence and perform independent verification	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
8	design code and validate an implementation plan	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
9	analyze a dataset and summarize the main findings	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
10	plan research and verify the resulting recommendations	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
11	inspect code and identify correctness risks	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
12	research alternatives and create a concise comparison	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
13	retrieve technical facts and verify consistency	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
14	analyze an architecture and propose an implementation plan	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
15	summarize multiple analyses and resolve contradictions	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
16	plan a system and verify its operational constraints	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
17	research a topic then code a small validated prototype	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
18	analyze evidence and produce a structured summary	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
19	retrieve information, plan next steps, and verify assumptions	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
20	code a solution then analyze and summarize its tradeoffs	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
21	research, analyze, verify and summarize a complex technical question	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
22	plan, code and verify a robust implementation	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
23	retrieve, analyze and validate evidence from independent sources	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
24	summarize findings while checking for contradictory claims	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
25	analyze a system for reliability and produce a remediation plan	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
26	research implementation options and verify scalability assumptions	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
27	retrieve evidence, analyze risks and summarize recommendations	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
28	plan an experiment, analyze expected outcomes and verify constraints	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
29	code and verify a data-processing workflow	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
30	research a protocol, analyze compatibility and summarize gaps	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
31	analyze a graph problem and plan an efficient solution	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
32	retrieve domain evidence and verify a proposed design	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
33	summarize a complex requirement and plan implementation steps	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
34	research multiple options, analyze costs and verify conclusions	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
35	code a prototype, analyze performance and summarize results	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
36	retrieve documentation, analyze behavior and verify compliance	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
37	plan a migration and analyze reliability risks	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
38	research, retrieve and verify supporting evidence	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
39	analyze competing technical claims and resolve disagreement	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.
40	summarize evidence, verify citations and plan follow-up work	AgentWeave Routing Benchmark v1	Domain-agnostic multi-capability requirements for routing and team-formation evaluation.

AgentWeave Tool-Routing Benchmark

Benchmark cases and evaluation artifacts for AgentWeave, a deterministic pre-inference routing layer for tool-rich language-model systems.

data/benchmark_cases.jsonl — the canonical, normalized Dataset Viewer split.
raw/evaluation/ — frozen heterogeneous evaluation artifacts retained as reproducibility evidence and intentionally excluded from the Viewer schema.
docs/ — BFCL result summaries and paper notes.

The explicit configs.data_files declaration above ensures Hugging Face renders only the normalized benchmark table in Dataset Viewer while preserving raw evaluation JSON files for download and audit.

Paper

AgentWeave: Routing Before Reasoning for Efficient Function Calling in Tool-Rich Language Models
arXiv: https://arxiv.org/abs/2608.23078

Related artifacts

Model/router artifact: https://huggingface.co/sauravsingla08/AgentWeave-Router-MiniLM
Interactive Space: https://huggingface.co/spaces/sauravsingla08/AgentWeave
Source code: https://github.com/sauravsingla/agentweave

Notes

This Hub repository publishes benchmark/evaluation data rather than model weights. The AgentWeave runtime and implementation remain in the canonical GitHub repository.