CSTutorBench
Models tutor students who use block-based programming to control a virtual robot. A rubric grades the feedback on 17 scenarios, including debugging and answer disclosure.
- Released
- July 2026
- Kind
- Other professions
- Level
- University
- Scored
- Graded by a model
- Answers
- Open-ended
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- Anyone
- Occupations
- Educational instruction and library
- Strongest model tested
-
Gemma 4 31B IT #101 on the capability index
- Last model added
- April 2026
- Scores refreshed
- November 20th, 2026
Gemma 4 31B IT leads with 0.89 among models released by April 2026 Gemma 4 31B IT leads with 0.89 among models released by April 2026
Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. This view shows only models we track, each at its best.
The newest model with a score here came out in April 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 1
Gemma 4 31B IT Trial 2 system prompt 0.89
- 2
Qwen3.5-9B Trial 2 system prompt 0.79
- 3 qwen3.6_latest Trial 2 system prompt 0.78
- 4 deepseek-r1_8b Trial 2 system prompt 0.77
- 5
Nemotron 3 Super Trial 2 system prompt 0.73
- 6
Gemma 4 E4B Trial 2 system prompt 0.72
- 7
Gemma 3 27B Trial 2 system prompt 0.69
- 8 gpt-oss-20b Trial 2 system prompt 0.64
- 9
Qwen3-Coder-30B-A3B-Instruct Trial 1 system prompt 0.53
- 10
Gemma 3 4B Trial 2 system prompt 0.50
- 11 olmo-3_7b-think Trial 2 system prompt 0.49
Source: Official CSTutorBench leaderboard
- 1
Gemma 4 31B IT Trial 2 system prompt 0.89
- 2
Qwen3.5-9B Trial 2 system prompt 0.79
- 3
Nemotron 3 Super Trial 2 system prompt 0.73
- 4
Gemma 4 E4B Trial 2 system prompt 0.72
- 5
Gemma 3 27B Trial 2 system prompt 0.69
- 6 gpt-oss-20b Trial 2 system prompt 0.64
- 7
Qwen3-Coder-30B-A3B-Instruct Trial 1 system prompt 0.53
- 8
Gemma 3 4B Trial 2 system prompt 0.50
Source: Official CSTutorBench leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 0.89 | Trial 2 system prompt | Apr 2026 | Official leaderboard |
| 2 | | 0.79 | Trial 2 system prompt | Feb 2026 | Official leaderboard |
| 3 | qwen3.6_latest | 0.78 | Trial 2 system prompt | Official leaderboard | |
| 4 | deepseek-r1_8b | 0.77 | Trial 2 system prompt | Official leaderboard | |
| 5 | | 0.73 | Trial 2 system prompt | Mar 2026 | Official leaderboard |
| 6 | | 0.72 | Trial 2 system prompt | Mar 2026 | Official leaderboard |
| 7 | | 0.69 | Trial 2 system prompt | Mar 2025 | Official leaderboard |
| 8 | gpt-oss-20b | 0.64 | Trial 2 system prompt | Aug 2025 | Official leaderboard |
| 9 | | 0.53 | Trial 1 system prompt | Jul 2025 | Official leaderboard |
| 10 | | 0.50 | Trial 2 system prompt | Mar 2025 | Official leaderboard |
| 11 | olmo-3_7b-think | 0.49 | Trial 2 system prompt | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 0.89 | Trial 2 system prompt | Apr 2026 | Official leaderboard |
| 2 | | 0.79 | Trial 2 system prompt | Feb 2026 | Official leaderboard |
| 3 | | 0.73 | Trial 2 system prompt | Mar 2026 | Official leaderboard |
| 4 | | 0.72 | Trial 2 system prompt | Mar 2026 | Official leaderboard |
| 5 | | 0.69 | Trial 2 system prompt | Mar 2025 | Official leaderboard |
| 6 | gpt-oss-20b | 0.64 | Trial 2 system prompt | Aug 2025 | Official leaderboard |
| 7 | | 0.53 | Trial 1 system prompt | Jul 2025 | Official leaderboard |
| 8 | | 0.50 | Trial 2 system prompt | Mar 2025 | Official leaderboard |
Rankings by part
We read 11 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as Conciseness, Vocabulary and Accuracy. Gemma 4 31B IT leads 4 of them, and 4 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
Conciseness
8 models we track have a score here.
- 1
Gemma 4 31B IT 1.00
- 2
Gemma 4 E4B ↑2 0.65
- 3
Qwen3.5-9B ↓1 0.62
- 4 gpt-oss-20b ↑2 0.41
- 5
Nemotron 3 Super ↓2 0.32
- 6
Gemma 3 27B ↓1 0.26
- 7
Qwen3-Coder-30B-A3B-Instruct 0.21
- 8
Gemma 3 4B 0.09
Vocabulary
8 models we track have a score here.
- 1
Gemma 3 27B ↑4 1.00
- 2
Gemma 3 4B ↑6 1.00
- 3
Qwen3-Coder-30B-A3B-Instruct ↑4 1.00
- 4
Qwen3.5-9B ↓2 1.00
- 5
Nemotron 3 Super ↓2 1.00
- 6
Gemma 4 31B IT ↓5 1.00
- 7 gpt-oss-20b ↓1 0.97
- 8
Gemma 4 E4B ↓4 0.94
Accuracy
8 models we track have a score here.
- 1
Gemma 4 31B IT 1.00
- 2
Qwen3.5-9B 0.85
- 3
Nemotron 3 Super 0.85
- 4
Qwen3-Coder-30B-A3B-Instruct ↑3 0.82
- 5
Gemma 4 E4B ↓1 0.79
- 6
Gemma 3 27B ↓1 0.76
- 7 gpt-oss-20b ↓1 0.76
- 8
Gemma 3 4B 0.47
Formatting
8 models we track have a score here.
- 1
Gemma 4 31B IT 1.00
- 2
Gemma 4 E4B ↑2 0.94
- 3
Qwen3.5-9B ↓1 0.91
- 4
Nemotron 3 Super ↓1 0.79
- 5
Gemma 3 27B 0.76
- 6 gpt-oss-20b 0.65
- 7
Gemma 3 4B ↑1 0.41
- 8
Qwen3-Coder-30B-A3B-Instruct ↓1 0.26
Tone
8 models we track have a score here.
- 1
Gemma 4 31B IT 0.94
- 2
Gemma 3 27B ↑3 0.91
- 3
Gemma 4 E4B ↑1 0.91
- 4
Qwen3.5-9B ↓2 0.88
- 5
Gemma 3 4B ↑3 0.79
- 6
Qwen3-Coder-30B-A3B-Instruct ↑1 0.74
- 7
Nemotron 3 Super ↓4 0.74
- 8 gpt-oss-20b ↓2 0.71
Actionability
8 models we track have a score here.
- 1
Nemotron 3 Super ↑2 0.82
- 2
Gemma 4 31B IT ↓1 0.79
- 3 gpt-oss-20b ↑3 0.74
- 4
Qwen3.5-9B ↓2 0.68
- 5
Gemma 4 E4B ↓1 0.59
- 6
Qwen3-Coder-30B-A3B-Instruct ↑1 0.53
- 7
Gemma 3 27B ↓2 0.50
- 8
Gemma 3 4B 0.21
Targetedness
8 models we track have a score here.
- 1
Qwen3.5-9B ↑1 0.85
- 2
Gemma 3 27B ↑3 0.79
- 3
Qwen3-Coder-30B-A3B-Instruct ↑4 0.79
- 4
Gemma 4 31B IT ↓3 0.79
- 5
Gemma 3 4B ↑3 0.74
- 6
Nemotron 3 Super ↓3 0.74
- 7 gpt-oss-20b ↓1 0.71
- 8
Gemma 4 E4B ↓4 0.71
Hint Not Solution
8 models we track have a score here.
- 1
Gemma 3 27B ↑4 0.81
- 2
Nemotron 3 Super ↑1 0.81
- 3
Gemma 4 31B IT ↓2 0.81
- 4
Gemma 4 E4B 0.75
- 5
Qwen3.5-9B ↓3 0.69
- 6
Gemma 3 4B ↑2 0.56
- 7
Qwen3-Coder-30B-A3B-Instruct 0.44
- 8 gpt-oss-20b ↓2 0.06
Acknowledges Progression
8 models we track have a score here.
- 1
Qwen3.5-9B ↑1 0.67
- 2
Gemma 4 31B IT ↓1 0.58
- 3
Gemma 3 27B ↑2 0.50
- 4
Qwen3-Coder-30B-A3B-Instruct ↑3 0.33
- 5
Nemotron 3 Super ↓2 0.33
- 6
Gemma 4 E4B ↓2 0.33
- 7
Gemma 3 4B ↑1 0.17
- 8 gpt-oss-20b ↓2 0.17
Builds On Success
8 models we track have a score here.
- 1
Qwen3.5-9B ↑1 1.00
- 2
Gemma 3 27B ↑3 0.50
- 3
Gemma 3 4B ↑5 0.50
- 4
Gemma 4 31B IT ↓3 0.50
- 5
Qwen3-Coder-30B-A3B-Instruct ↑2 0.00
- 6 gpt-oss-20b 0.00
- 7
Nemotron 3 Super ↓4 0.00
- 8
Gemma 4 E4B ↓4 0.00
Conceptual Clarity
8 models we track have a score here.
- 1 gpt-oss-20b ↑5 1.00
- 2
Gemma 3 27B ↑3 0.75
- 3
Qwen3.5-9B ↓1 0.75
- 4
Nemotron 3 Super ↓1 0.75
- 5
Gemma 4 31B IT ↓4 0.75
- 6
Qwen3-Coder-30B-A3B-Instruct ↑1 0.50
- 7
Gemma 4 E4B ↓3 0.50
- 8
Gemma 3 4B 0.00
Scores at each effort level
Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.
Source: Official CSTutorBench leaderboard
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| Trial 2 system prompt | 0.89 | ||||
| Trial 2 system prompt | 0.79 | ||||
| Trial 1 system prompt | 0.77 | ||||
| Trial 2 system prompt | 0.73 | ||||
| Trial 2 system prompt | 0.72 | ||||
| Trial 2 system prompt | 0.69 | ||||
| Trial 1 system prompt | 0.66 | ||||
| Trial 1 system prompt | 0.65 | ||||
| gpt-oss-20b | Trial 2 system prompt | 0.64 | |||
| Trial 1 system prompt | 0.58 | ||||
| Trial 1 system prompt | 0.57 | ||||
| gpt-oss-20b | Trial 1 system prompt | 0.55 | |||
| Trial 1 system prompt | 0.53 | ||||
| Trial 2 system prompt | 0.52 | ||||
| Trial 2 system prompt | 0.50 | ||||
| Trial 1 system prompt | 0.36 |
Scores at each effort level
Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.
Source: Official CSTutorBench leaderboard
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| Trial 2 system prompt | 0.89 | ||||
| Trial 2 system prompt | 0.79 | ||||
| Trial 1 system prompt | 0.77 | ||||
| Trial 2 system prompt | 0.73 | ||||
| Trial 2 system prompt | 0.72 | ||||
| Trial 2 system prompt | 0.69 | ||||
| Trial 1 system prompt | 0.66 | ||||
| Trial 1 system prompt | 0.65 | ||||
| gpt-oss-20b | Trial 2 system prompt | 0.64 | |||
| Trial 1 system prompt | 0.58 | ||||
| Trial 1 system prompt | 0.57 | ||||
| gpt-oss-20b | Trial 1 system prompt | 0.55 | |||
| Trial 1 system prompt | 0.53 | ||||
| Trial 2 system prompt | 0.52 | ||||
| Trial 2 system prompt | 0.50 | ||||
| Trial 1 system prompt | 0.36 |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official CSTutorBench leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official CSTutorBench leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Law, finance and medicine
- MedXpertQA-MM In use Models answer expert-level medical exam questions that combine images with clinical records and test results. Scoring is accuracy.
Qwen 3.8 Max 80.4%
- HealthBench Professional In use 525 physician-authored tasks from real clinician chats (care consult, documentation, research) with physician-adjudicated rubrics. Claude Sonnet 5.5 69.2%
- HealthBench In use 5,000 realistic health conversations graded against physician-written rubrics. A length-adjusted variant penalizes verbosity. GPT-5 67.2%
- FinSearchComp In use Agents search the web to fetch live financial data, look up history, and run multi-step investigations, across 635 expert-written questions. Scored by answer accuracy.
Ling-3.1-flash 81.7%
- KorMedMCQA In use Models answer 7,469 multiple-choice questions from Korean licensing exams for doctors, nurses, pharmacists and dentists, held from 2012 to 2024. o1-preview 92.7%
- Harvey LAB-AA v1.1 In use Artificial Analysis's run of Harvey's Legal Agent Benchmark on Harvey's newer private set of 120 tasks across 24 practice areas. A deliverable with a material hallucination scores zero, and the headline is the share of judges who find every rubric criterion passed. GPT-6.2 Astra 15%
Data audit
8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 7 months ago, in April 2026, for Gemma 4 31B IT. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
- Pass Independent runs Scores from the official leaderboard (11 entries).
- Pass Models scored 8 models we track, plus 3 other entries on its leaderboard.
- Warning Newest score The newest came 7 months ago, in April 2026, for Gemma 4 31B IT.
- Note Frontier models None of the top 10 has a score. The strongest tested is Gemma 4 31B IT (#101).
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on github.com: 16 rows for models we track and 3 other entries at the last read.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.