# TutorBench

> Source: Benchize (https://benchize.com) — https://benchize.com/benchmarks/scale%3Atutorbench. When citing or reusing this content, credit "Benchize" and link to https://benchize.com/benchmarks/scale%3Atutorbench.

Large Language Models serve as on-demand tutors for learners worldwide, yet, a critical evaluation gap exists. While most benchmarks assess an LLM's ability to solve problems, this capability alone does not make the models effective tutors. Effective tutors require nuanced, human-centered skills essential for student learning like providing adaptive explanations, offering guiding feedback, and adjusting to a learner's specific needs. To address this gap, we introduce TutorBench, a comprehensive benchmark designed to rigorously evaluate the core tutoring skills of LLMs. TutorBench moves beyond simple answer-correctness to measure how well models perform three common and critical tutoring tasks:

Canonical URL: https://benchize.com/benchmarks/scale%3Atutorbench
Last updated: Wed, 21 Oct 2026 18:52:20 GMT

Reproduction and quotation must name Benchize, link to the canonical HTML URL, and retain individual source citations.

Large Language Models serve as on-demand tutors for learners worldwide, yet, a critical evaluation gap exists. While most benchmarks assess an LLM's ability to solve problems, this capability alone does not make the models effective tutors. Effective tutors require nuanced, human-centered skills essential for student learning like providing adaptive explanations, offering guiding feedback, and adjusting to a learner's specific needs. To address this gap, we introduce TutorBench, a comprehensive benchmark designed to rigorously evaluate the core tutoring skills of LLMs. TutorBench moves beyond simple answer-correctness to measure how well models perform three common and critical tutoring tasks:

Professional work · Models tested: 5 · Score publishers: [scale](https://labs.scale.com/leaderboard/tutorbench)

## Model results

| # | Model | Score | ± |  |
| --- | --- | --- | --- | --- |
| 1 | [Gemini 2.5 Pro Preview 06-05](https://benchize.com/models/gemini-2-5-pro-preview.md) | 55.65source score | CI 54.54–56.76 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 55.65 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration gemini-2.5-pro-preview-06-05 sourceModelVersion Not reported thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 54.54–56.76 contaminationWarning Not reported rank 3 reportedConfidenceHalfWidth 1.11 ### Source [Original source](https://labs.scale.com/leaderboard/tutorbench) Source scale Original model label gemini-2.5-pro-preview-06-05 Observed 2026-10-21T16:35:39.395Z Evidence SHA-256 `3b52113d689cb5847cb39e84b957de8a2eab854c12eef25eedee4751268a6361` Source locator `/rows/2` |  |  |  |  |
| 2 | [Claude Opus 4.6](https://benchize.com/models/claude-opus-4-6-off.md)Non-thinking | 53.55source score | CI 52.54–54.56 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 53.55 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort non-thinking Model configuration claude-opus-4-6 (Non-Thinking) sourceModelVersion Not reported thinkingEnabled false ### Reported statistics Uncertainty Confidence interval: 52.54–54.56 contaminationWarning Not reported rank 10 reportedConfidenceHalfWidth 1.01 ### Source [Original source](https://labs.scale.com/leaderboard/tutorbench) Source scale Original model label claude-opus-4-6 (Non-Thinking) Observed 2026-10-21T16:35:39.395Z Evidence SHA-256 `3b52113d689cb5847cb39e84b957de8a2eab854c12eef25eedee4751268a6361` Source locator `/rows/9` |  |  |  |  |
| 3 | [Gemini 3.1 Pro Preview](https://benchize.com/models/gemini-3-1-pro-preview.md) | 52.99source score | CI 51.93–54.05 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 52.9926 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration gemini-3.1-pro-preview sourceModelVersion Not reported thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 51.93–54.05 contaminationWarning Not reported rank 12 reportedConfidenceHalfWidth 1.06 ### Source [Original source](https://labs.scale.com/leaderboard/tutorbench) Source scale Original model label gemini-3.1-pro-preview Observed 2026-10-21T16:35:39.395Z Evidence SHA-256 `3b52113d689cb5847cb39e84b957de8a2eab854c12eef25eedee4751268a6361` Source locator `/rows/11` |  |  |  |  |
| 4 | [Gemini 3.1 Flash Lite Preview](https://benchize.com/models/gemini-3-1-flash-lite-preview.md) | 51.5source score | CI 50.46–52.54 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 51.5049 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration gemini-3.1-flash-lite-preview sourceModelVersion Not reported thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 50.46–52.54 contaminationWarning Not reported rank 15 reportedConfidenceHalfWidth 1.04 ### Source [Original source](https://labs.scale.com/leaderboard/tutorbench) Source scale Original model label gemini-3.1-flash-lite-preview Observed 2026-10-21T16:35:39.395Z Evidence SHA-256 `3b52113d689cb5847cb39e84b957de8a2eab854c12eef25eedee4751268a6361` Source locator `/rows/14` |  |  |  |  |
| 5 | [GPT-4o](https://benchize.com/models/gpt-4o.md) | 36.12source score | CI 35.16–37.08 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 36.12 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration gpt-4o sourceModelVersion Not reported thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 35.16–37.08 contaminationWarning Not reported rank 27 reportedConfidenceHalfWidth 0.96 ### Source [Original source](https://labs.scale.com/leaderboard/tutorbench) Source scale Original model label gpt-4o Observed 2026-10-21T16:35:39.395Z Evidence SHA-256 `3b52113d689cb5847cb39e84b957de8a2eab854c12eef25eedee4751268a6361` Source locator `/rows/26` |  |  |  |  |

---
Cite as: TutorBench — Benchize, https://benchize.com/benchmarks/scale%3Atutorbench (retrieved 2026-10-21)