EduBench-RU
Models plan lessons, explain school subjects, and assist teachers in Russian and Chuvash. A panel of model judges scores responses against a teaching rubric.
- Made by
- Daniel Ivanov and Alexander Antonov
- Released
- March 2026
- Kind
- Other professions
- Level
- University
- Scored
- Graded by a model
- Answers
- Open-ended
- Tasks
- All public
- Measures
- Capability
- Subject
- Education
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- Anyone
- Occupations
- Educational instruction and library
- Strongest model tested
- GPT-5.4 #28 on the capability index
- Last model added
- March 2026
- Scores refreshed
- November 20th, 2026
Gemini 3.1 Pro leads with 3.61 among models released by March 2026 Gemini 3.1 Pro leads with 3.61 among models released by March 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. 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 March 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
- 3.61
Gemini 3.1 Pro
- 3.40
Gemini 3.1 Flash-Lite
- 3.36 Claude Opus 4.6
- 3.34
Gemini 2.5 Pro
- 3.33 Claude Sonnet 4.6
- 3.24
DeepSeek-V3.2
- 3.24
Mistral Large 3
- 3.22 Kimi K2.5
- 3.21 EduLLM-RU 27B Q6_K (ours)
- 3.20 GLM-5
- 3.20 GPT-5.4
- 3.07 GPT-5.4 Mini
- 3.01
Gemini 2.5 Flash
- 3.00 Qwen3.5 27B (base)
- 2.94 Grok 4.1 Fast
- 2.83
Qwen3-235B-A22B
- 2.69 EduLLM-RU 32B Q6_K (ours)
- 2.68 glm-4.7-flash
- 2.61
Qwen3-32B
- 2.58
Llama 4 Maverick
- 2.51 YandexGPT 5.1 Pro
- 2.41
Qwen3-14B
- 2.39 GigaChat-2 Max
- 2.36 GigaChat-2 Pro
- 2.34
Qwen3-8B
- 2.27 YandexGPT 5 Lite
- 2.23
Cohere Command A
- 2.16 GigaChat-2 Lite
- 1.57
Phi-4
- 1
Gemini 3.1 Pro Cloud 3.61
- 2
Gemini 3.1 Flash-Lite Cloud 3.40
- 3 Claude Opus 4.6 Cloud 3.36
- 4
Gemini 2.5 Pro Jun 2025, Cloud 3.34
- 5 Claude Sonnet 4.6 Cloud 3.33
- 6
DeepSeek-V3.2 Cloud 3.24
- 7
Mistral Large 3 Open 3.24
- 8 Kimi K2.5 Cloud 3.22
- 9 EduLLM-RU 27B Q6_K (ours) Self-hosted 3.21
- 10 GLM-5 Cloud 3.20
- 11 GPT-5.4 Cloud 3.20
- 12 GPT-5.4 Mini Cloud 3.07
- 13
Gemini 2.5 Flash Sep 2025, Cloud 3.01
- 14 Qwen3.5 27B (base) Open 3.00
- 15 Grok 4.1 Fast Cloud 2.94
- 16
Qwen3-235B-A22B Open 2.83
- 17 EduLLM-RU 32B Q6_K (ours) Self-hosted 2.69
- 18 glm-4.7-flash Cloud 2.68
- 19
Qwen3-32B Open 2.61
- 20
Llama 4 Maverick Open 2.58
- 21 YandexGPT 5.1 Pro Russian cloud 2.51
- 22
Qwen3-14B Open 2.41
- 23 GigaChat-2 Max Russian cloud 2.39
- 24 GigaChat-2 Pro Russian cloud 2.36
- 25
Qwen3-8B Open 2.34
- 26 YandexGPT 5 Lite Russian cloud 2.27
- 27
Cohere Command A Cloud 2.23
- 28 GigaChat-2 Lite Russian cloud 2.16
- 29
Phi-4 Open 1.57
Source: Official EduBench-RU leaderboard
- 3.61
Gemini 3.1 Pro
- 3.40
Gemini 3.1 Flash-Lite
- 3.36 Claude Opus 4.6
- 3.34
Gemini 2.5 Pro
- 3.33 Claude Sonnet 4.6
- 3.24
DeepSeek-V3.2
- 3.24
Mistral Large 3
- 3.22 Kimi K2.5
- 3.20 GLM-5
- 3.20 GPT-5.4
- 3.07 GPT-5.4 Mini
- 3.01
Gemini 2.5 Flash
- 2.94 Grok 4.1 Fast
- 2.83
Qwen3-235B-A22B
- 2.68 glm-4.7-flash
- 2.61
Qwen3-32B
- 2.58
Llama 4 Maverick
- 2.41
Qwen3-14B
- 2.34
Qwen3-8B
- 2.23
Cohere Command A
- 1.57
Phi-4
- 1
Gemini 3.1 Pro Cloud 3.61
- 2
Gemini 3.1 Flash-Lite Cloud 3.40
- 3 Claude Opus 4.6 Cloud 3.36
- 4
Gemini 2.5 Pro Jun 2025, Cloud 3.34
- 5 Claude Sonnet 4.6 Cloud 3.33
- 6
DeepSeek-V3.2 Cloud 3.24
- 7
Mistral Large 3 Open 3.24
- 8 Kimi K2.5 Cloud 3.22
- 9 GLM-5 Cloud 3.20
- 10 GPT-5.4 Cloud 3.20
- 11 GPT-5.4 Mini Cloud 3.07
- 12
Gemini 2.5 Flash Sep 2025, Cloud 3.01
- 13 Grok 4.1 Fast Cloud 2.94
- 14
Qwen3-235B-A22B Open 2.83
- 15 glm-4.7-flash Cloud 2.68
- 16
Qwen3-32B Open 2.61
- 17
Llama 4 Maverick Open 2.58
- 18
Qwen3-14B Open 2.41
- 19
Qwen3-8B Open 2.34
- 20
Cohere Command A Cloud 2.23
- 21
Phi-4 Open 1.57
Source: Official EduBench-RU leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 3.61 | Cloud | Feb 2026 | Official leaderboard |
| 2 | | 3.40 | Cloud | Mar 2026 | Official leaderboard |
| 3 | Claude Opus 4.6 | 3.36 | Cloud | Feb 2026 | Official leaderboard |
| 4 | | 3.34 | Cloud | Jun 2025 | Official leaderboard |
| 5 | Claude Sonnet 4.6 | 3.33 | Cloud | Feb 2026 | Official leaderboard |
| 6 | | 3.24 | Cloud | Dec 2025 | Official leaderboard |
| 7 | | 3.24 | Open | Dec 2025 | Official leaderboard |
| 8 | Kimi K2.5 | 3.22 | Cloud | Jan 2026 | Official leaderboard |
| 9 | EduLLM-RU 27B Q6_K (ours) | 3.21 | Self-hosted | Official leaderboard | |
| 10 | GLM-5 | 3.20 | Cloud | Feb 2026 | Official leaderboard |
| 11 | GPT-5.4 | 3.20 | Cloud | Mar 2026 | Official leaderboard |
| 12 | GPT-5.4 Mini | 3.07 | Cloud | Mar 2026 | Official leaderboard |
| 13 | | 3.01 | Cloud | Sep 2025 | Official leaderboard |
| 14 | Qwen3.5 27B (base) | 3.00 | Open | Official leaderboard | |
| 15 | Grok 4.1 Fast | 2.94 | Cloud | Nov 2025 | Official leaderboard |
| 16 | | 2.83 | Open | Apr 2025 | Official leaderboard |
| 17 | EduLLM-RU 32B Q6_K (ours) | 2.69 | Self-hosted | Official leaderboard | |
| 18 | glm-4.7-flash | 2.68 | Cloud | Jan 2026 | Official leaderboard |
| 19 | | 2.61 | Open | Apr 2025 | Official leaderboard |
| 20 | | 2.58 | Open | Apr 2025 | Official leaderboard |
| 21 | YandexGPT 5.1 Pro | 2.51 | Russian cloud | Official leaderboard | |
| 22 | | 2.41 | Open | Apr 2025 | Official leaderboard |
| 23 | GigaChat-2 Max | 2.39 | Russian cloud | Official leaderboard | |
| 24 | GigaChat-2 Pro | 2.36 | Russian cloud | Official leaderboard | |
| 25 | | 2.34 | Open | Apr 2025 | Official leaderboard |
| 26 | YandexGPT 5 Lite | 2.27 | Russian cloud | Official leaderboard | |
| 27 | | 2.23 | Cloud | Mar 2025 | Official leaderboard |
| 28 | GigaChat-2 Lite | 2.16 | Russian cloud | Official leaderboard | |
| 29 | | 1.57 | Open | Dec 2024 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 3.61 | Cloud | Feb 2026 | Official leaderboard |
| 2 | | 3.40 | Cloud | Mar 2026 | Official leaderboard |
| 3 | Claude Opus 4.6 | 3.36 | Cloud | Feb 2026 | Official leaderboard |
| 4 | | 3.34 | Cloud | Jun 2025 | Official leaderboard |
| 5 | Claude Sonnet 4.6 | 3.33 | Cloud | Feb 2026 | Official leaderboard |
| 6 | | 3.24 | Cloud | Dec 2025 | Official leaderboard |
| 7 | | 3.24 | Open | Dec 2025 | Official leaderboard |
| 8 | Kimi K2.5 | 3.22 | Cloud | Jan 2026 | Official leaderboard |
| 9 | GLM-5 | 3.20 | Cloud | Feb 2026 | Official leaderboard |
| 10 | GPT-5.4 | 3.20 | Cloud | Mar 2026 | Official leaderboard |
| 11 | GPT-5.4 Mini | 3.07 | Cloud | Mar 2026 | Official leaderboard |
| 12 | | 3.01 | Cloud | Sep 2025 | Official leaderboard |
| 13 | Grok 4.1 Fast | 2.94 | Cloud | Nov 2025 | Official leaderboard |
| 14 | | 2.83 | Open | Apr 2025 | Official leaderboard |
| 15 | glm-4.7-flash | 2.68 | Cloud | Jan 2026 | Official leaderboard |
| 16 | | 2.61 | Open | Apr 2025 | Official leaderboard |
| 17 | | 2.58 | Open | Apr 2025 | Official leaderboard |
| 18 | | 2.41 | Open | Apr 2025 | Official leaderboard |
| 19 | | 2.34 | Open | Apr 2025 | Official leaderboard |
| 20 | | 2.23 | Cloud | Mar 2025 | Official leaderboard |
| 21 | | 1.57 | Open | Dec 2024 | Official leaderboard |
Rankings by part
We read 2 parts from the benchmark’s own leaderboard: EduRU and ChvBn. Each ranks models on one part of the test. Gemini 3.1 Pro leads both. Arrows show how far a model moved from its overall rank among the models we track.
EduRU
21 models we track have a score here.
- 1
Gemini 3.1 Pro 3.64
- 2
Gemini 3.1 Flash-Lite 3.51
- 3 Claude Opus 4.6 3.46
- 4 Claude Sonnet 4.6 ↑1 3.43
- 5
Gemini 2.5 Pro Jun 2025 ↓1 3.38
- 6
DeepSeek-V3.2 3.37
- 7
Mistral Large 3 3.35
- 8 Kimi K2.5 3.35
- 9 GLM-5 3.34
- 10 GPT-5.4 3.33
ChvBn
21 models we track have a score here.
- 1
Gemini 3.1 Pro 3.43
- 2
Gemini 2.5 Pro Jun 2025 ↑2 3.16
- 3
Gemini 3.1 Flash-Lite ↓1 3.02
- 4 Claude Opus 4.6 ↓1 3.00
- 5 Claude Sonnet 4.6 3.00
- 6
Gemini 2.5 Flash Sep 2025 ↑6 2.86
- 7
Mistral Large 3 2.82
- 8 GPT-5.4 ↑2 2.82
- 9
DeepSeek-V3.2 ↓3 2.77
- 10 GLM-5 ↓1 2.77
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official EduBench-RU 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 EduBench-RU 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 8 months ago, in March 2026, for GPT-5.4 Mini. 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 (29 entries).
- Pass Models scored 21 models we track, plus 8 other entries on its leaderboard.
- Warning Newest score The newest came 8 months ago, in March 2026, for GPT-5.4 Mini.
- Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5.4 (#28).
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on github.com: 21 rows for models we track and 8 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.