# RelyToolBench

> Tool-hallucination benchmark that checks whether a model selects and calls tools that fit the request.

Canonical page: https://benchlm.ai/benchmarks/relytoolbench

- Category: [Agentic](/agentic)
- Last updated: October 29, 2026

## About RelyToolBench

- Year: 2025
- Tasks: Tool-use requests
- Format: Pass rate
- Difficulty: Tool grounding and hallucination
- Paper: [RelyToolBench](https://github.com/X-LANCE/ToolHallucination)

NVIDIA reports it for Nemotron 3.5 Bolt E4B (October 23, 2026), run in its NeMo Gym harness, as the native/real pass rate in reasoning mode.

RelyToolBench is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (2 models)

| Rank | Model | Creator | Score |
|------|-------|---------|-------|
| 1 | [Nemotron 3.5 Bolt E4B](/models/nemotron-3-5-bolt-e4b) | NVIDIA | 69.29% |
| 2 | [Nemotron 3.5 Bolt E4B BF16](/models/nemotron-3-5-bolt-e4b-bf16) | NVIDIA | 63.16% |

## FAQ

### What does RelyToolBench measure?

Tool-hallucination benchmark that checks whether a model selects and calls tools that fit the request.

### Which model scores highest on RelyToolBench?

Nemotron 3.5 Bolt E4B by NVIDIA currently leads with a score of 69.29% on RelyToolBench.

### How many models are evaluated on RelyToolBench?

2 AI models have published results on RelyToolBench in the BenchLM catalog.

### Does RelyToolBench affect BenchLM's overall score?

Not directly. RelyToolBench is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Compare Top Models on RelyToolBench

- [Nemotron 3.5 Bolt E4B vs Nemotron 3.5 Bolt E4B BF16](/compare/nemotron-3-5-bolt-e4b-vs-nemotron-3-5-bolt-e4b-bf16)