TMMLU+ 1.1
Version 1.1
Models answer traditional Chinese subject questions from a revised exam set. The revision corrects 197 answers and removes 539 questions from the original release.
The authors recomputed results on the 22,203 remaining questions.
- Made by
- iKala
- Released
- September 2026
- Kind
- Knowledge in many languages
- Level
- University
- Scored
- Checked automatically
- Answers
- Multiple choice
- Tasks
- All public
- Measures
- Capability
- Language
- Chinese
- Interaction
- Single turn
- Items from
- Another benchmark, Exams and contests
- Refresh
- New versions replace it
- Run by
- Anyone
- Based on
- TMMLU+
- In views
- Biology and medicine, Law and finance
- Strongest model tested
- Claude Opus 5 #11 on the capability index
- Last model added
- August 2026
- Scores refreshed
- November 20th, 2026
Claude Opus 5 leads with 96.2% among models released by August 2026
Best score per model from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best.
The newest model with a score here came out in August 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 12 more
- 96.2 Claude Opus 5
- 90.5
Gemini 3.7 Flash
- 88.4 GPT-5.6 Sol
- 87.1 Claude Fable 5
- 86.4
DeepSeek V4 Pro 0813
- 84.7 Claude Sonnet 5
- 84.3
Gemini 3.1 Pro
- 83.5
DeepSeek-V4-Flash
- 83.3
Tencent Hy3
- 82.4 GPT-5.6 Terra
- 81.0 Kimi K3
- 80.6
Qwen3.7-Max
- 80.0
Qwen 3.8 27B
- 79.6 GPT-5.6 Luna
- 78.6 GLM-5.2
- 78.1 MiMo-V2.5
- 76.4 Grok 4.3
- 76.4
MiniMax-M3
- 71.7 Claude Haiku 4.5
- 69.8
Gemma 4 31B IT
- 67.5
Nemotron 3 Ultra
- 1 Claude Opus 5 96.2%
- 2
Gemini 3.7 Flash reasoning 90.5%
- 3 GPT-5.6 Sol 88.4%
- 4 Claude Fable 5 87.1%
- 5
DeepSeek V4 Pro 0813 86.4%
- 6 Claude Sonnet 5 84.7%
- 7
Gemini 3.1 Pro reasoning 84.3%
- 8
DeepSeek-V4-Flash 83.5%
- 9
Tencent Hy3 83.3%
- 10 GPT-5.6 Terra 82.4%
- 11 Kimi K3 81%
- 12
Qwen3.7-Max 80.6%
- 13
Qwen 3.8 27B 80%
- 14 GPT-5.6 Luna 79.6%
- 15 GLM-5.2 78.6%
- 16 MiMo-V2.5 78.1%
- 17 Grok 4.3 76.4%
- 18
MiniMax-M3 76.4%
- 19 Claude Haiku 4.5 71.7%
- 20
Gemma 4 31B IT 69.8%
- 21
Nemotron 3 Ultra 67.5%
Source: Official TMMLU+ 1.1 leaderboard
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude Opus 5 | 96.2% | Jul 2026 | Official leaderboard | |
| 2 | | 90.5% | reasoning | Aug 2026 | Official leaderboard |
| 3 | GPT-5.6 Sol | 88.4% | Jul 2026 | Official leaderboard | |
| 4 | Claude Fable 5 | 87.1% | Jun 2026 | Official leaderboard | |
| 5 | | 86.4% | Aug 2026 | Official leaderboard | |
| 6 | Claude Sonnet 5 | 84.7% | Jun 2026 | Official leaderboard | |
| 7 | | 84.3% | reasoning | Feb 2026 | Official leaderboard |
| 8 | | 83.5% | Apr 2026 | Official leaderboard | |
| 9 | | 83.3% | Jul 2026 | Official leaderboard | |
| 10 | GPT-5.6 Terra | 82.4% | Jul 2026 | Official leaderboard | |
| 11 | Kimi K3 | 81% | Jul 2026 | Official leaderboard | |
| 12 | | 80.6% | May 2026 | Official leaderboard | |
| 13 | | 80% | Aug 2026 | Official leaderboard | |
| 14 | GPT-5.6 Luna | 79.6% | Jul 2026 | Official leaderboard | |
| 15 | GLM-5.2 | 78.6% | Jun 2026 | Official leaderboard | |
| 16 | MiMo-V2.5 | 78.1% | Apr 2026 | Official leaderboard | |
| 17 | Grok 4.3 | 76.4% | Apr 2026 | Official leaderboard | |
| 18 | | 76.4% | Jun 2026 | Official leaderboard | |
| 19 | Claude Haiku 4.5 | 71.7% | Oct 2025 | Official leaderboard | |
| 20 | | 69.8% | Apr 2026 | Official leaderboard | |
| 21 | | 67.5% | Jun 2026 | Official leaderboard |
Rankings by part
We read 4 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as STEM, Social Science and Humanities. Claude Opus 5 leads every one of them. Arrows show how far a model moved from its overall rank among the models we track.
STEM
21 models we track have a score here.
- 1 Claude Opus 5 98.7%
- 2
Gemini 3.7 Flash 93.4%
- 3 GPT-5.6 Sol 93.3%
- 4
Qwen 3.8 27B ↑9 93.1%
- 5
DeepSeek V4 Pro 0813 92.9%
- 6 Claude Fable 5 ↓2 90.6%
- 7 GPT-5.6 Terra ↑3 90.2%
- 8
Tencent Hy3 ↑1 89.8%
- 9
DeepSeek-V4-Flash ↓1 89.4%
- 10 Claude Sonnet 5 ↓4 89.2%
Social Science
21 models we track have a score here.
- 1 Claude Opus 5 96.1%
- 2
Gemini 3.7 Flash 92.1%
- 3 GPT-5.6 Sol 91.5%
- 4
DeepSeek V4 Pro 0813 ↑1 89.2%
- 5
Tencent Hy3 ↑4 88.3%
- 6 GPT-5.6 Terra ↑4 87%
- 7
DeepSeek-V4-Flash ↑1 86.7%
- 8 Claude Sonnet 5 ↓2 85.8%
- 9
Qwen3.7-Max ↑3 84%
- 10
Qwen 3.8 27B ↑3 84%
Humanities
21 models we track have a score here.
- 1 Claude Opus 5 94.6%
- 2 Claude Fable 5 ↑2 90%
- 3
Gemini 3.7 Flash ↓1 88.9%
- 4
Gemini 3.1 Pro ↑3 84.9%
- 5 GPT-5.6 Sol ↓2 82.3%
- 6 Claude Sonnet 5 81.7%
- 7
DeepSeek V4 Pro 0813 ↓2 79.6%
- 8
DeepSeek-V4-Flash 77.4%
- 9 Kimi K3 ↑2 77.3%
- 10
Qwen3.7-Max ↑2 76%
Other
21 models we track have a score here.
- 1 Claude Opus 5 95.3%
- 2
Gemini 3.7 Flash 87.6%
- 3 GPT-5.6 Sol 86.4%
- 4 Claude Fable 5 84.2%
- 5
DeepSeek V4 Pro 0813 83.9%
- 6 Claude Sonnet 5 81.9%
- 7
Gemini 3.1 Pro 81.8%
- 8
Tencent Hy3 ↑1 80.5%
- 9
DeepSeek-V4-Flash ↓1 80.4%
- 10 GPT-5.6 Terra 80.1%
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official TMMLU+ 1.1 leaderboard
Claude Fable 5 on the official leaderboard passed 85% in June 2026
Models released before it came out already scored that high. All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official TMMLU+ 1.1 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.
Versions
We switch the top row to each new version when it comes out. Open it for one row per version. An old version stays in play while people still test it, so whoever first takes it past 85% gets credit for closing it out.
Source: Official leaderboards
More in Multilingual
- MMMLU In use MMLU professionally translated into 14 languages. Multiple-choice accuracy. Claude Mythos Preview 92.7%
- WMT24++ In use English-to-55-language translation of WMT24 test data. Scored with automatic metrics such as MetricX or xCOMET.
K-EXAONE 90.5%
- MMLU-ProX In use MMLU-Pro translated into up to 29 languages with expert review. Multiple-choice accuracy per language.
Qwen3.7-Max 87%
- CLIcK In use Models answer 1,995 multiple-choice questions on Korean language and culture. Questions come from official Korean exams and textbooks in eleven categories.
Solar Open 2 90.7%
- Global MMLU Lite 3.0 In use Models answer 400 human-translated MMLU questions per language in 23 languages. Version 3.0 adds Czech, Hungarian, Oriya, Slovak and Tajik and updates Italian. Accuracy. Claude Opus 5 95.7%
- MMMB In use Models answer multiple-choice questions about images in six languages across 15 categories. The set has about 12,000 questions. LFM2.5-VL-3B 83%
Data audit
All 9 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
- Pass Independent runs Scores from the official leaderboard (21 entries).
- Pass Models scored 21 models we track.
- Pass Newest score The newest is from August 2026, for Qwen 3.8 27B.
- Note Frontier models None of the top 10 has a score, which is fine for a retired test. The strongest tested is Claude Opus 5 (#11).
- Note Headroom Saturated. The best is 96.2%, past 85% since June 2026.
Leaderboard import 3 pass
- Pass Importer We read the board on huggingface.co: 21 rows for models we track 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.