Kaggle's Task Tuesday Benchmark
Models attempt tasks selected by the Kaggle team from community submissions. The board reports an overall score and individual task results.
- Made by
- Kaggle
- Released
- January 2026
- Kind
- Language reasoning
- Level
- University
- Scored
- Settled by events
- Answers
- Open-ended
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Single turn
- Items from
- Written by crowdworkers
- Refresh
- One fixed set
- Run by
- Anyone
- Strongest model tested
-
Gemini 3 Pro #56 on the capability index
- Last model added
- December 2025
- Scores refreshed
- November 20th, 2026
Gemini 3 Flash leads with 83.3 among models released by December 2025
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 December 2025, 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
Gemini 3 Flash 83.3
- 2 Claude Sonnet 4.5 75.0
- 3
Qwen3-Next-80B-A3B 66.7
- 4 Claude Opus 4.5 66.7
- 5
Gemini 3 Pro 57.1
- 6
DeepSeek-V3.2 50.0
Source: Official Kaggle's Task Tuesday Benchmark leaderboard
| # | Model | Score | Setting | Released | Run on | Source |
|---|---|---|---|---|---|---|
| 1 | | 83.3 | Dec 2025 | Feb 2026 | Official leaderboard | |
| 2 | Claude Sonnet 4.5 | 75.0 | Sep 2025 | Jan 2026 | Official leaderboard | |
| 3 | | 66.7 | Sep 2025 | Feb 2026 | Official leaderboard | |
| 4 | Claude Opus 4.5 | 66.7 | Nov 2025 | Feb 2026 | Official leaderboard | |
| 5 | | 57.1 | Nov 2025 | Mar 2026 | Official leaderboard | |
| 6 | | 50.0 | Dec 2025 | Feb 2026 | Official leaderboard |
Rankings by part
We read 12 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as Psycho Mirror, WWTP Confined Space Emergency Response and Matplotlib art. Qwen3-Next-80B-A3B leads 6 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.
Psycho Mirror
6 models we track have a score here.
- 1 Claude Opus 4.5 ↑3 0.76
- 2 Claude Sonnet 4.5 0.70
- 3
DeepSeek-V3.2 ↑3 0.66
- 4
Gemini 3 Flash ↓3 0.64
- 5
Gemini 3 Pro 0.64
- 6
Qwen3-Next-80B-A3B ↓3 0.57
WWTP Confined Space Emergency Response
6 models we track have a score here.
- 1
Qwen3-Next-80B-A3B ↑2 100.0
- 2 Claude Sonnet 4.5 100.0
- 3
Gemini 3 Pro ↑2 100.0
- 4 Claude Opus 4.5 100.0
- 5
Gemini 3 Flash ↓4 100.0
- 6
DeepSeek-V3.2 0.00
Matplotlib art
6 models we track have a score here.
- 1
Gemini 3 Pro ↑4 0.85
- 2
Gemini 3 Flash ↓1 0.84
- 3 Claude Opus 4.5 ↑1 0.83
- 4 Claude Sonnet 4.5 ↓2 0.53
- 5
DeepSeek-V3.2 ↑1 0.42
- 6
Qwen3-Next-80B-A3B ↓3 0.36
Rail Fence
6 models we track have a score here.
- 1
Qwen3-Next-80B-A3B ↑2 100.0
- 2 Claude Sonnet 4.5 100.0
- 3
Gemini 3 Pro ↑2 100.0
- 4 Claude Opus 4.5 100.0
- 5
DeepSeek-V3.2 ↑1 100.0
- 6
Gemini 3 Flash ↓5 100.0
spiral_neighbor_sum_hexagon
6 models we track have a score here.
- 1
Qwen3-Next-80B-A3B ↑2 100.0
- 2
Gemini 3 Flash ↓1 100.0
- 3 Claude Sonnet 4.5 ↓1 0.00
- 4
Gemini 3 Pro ↑1 0.00
- 5 Claude Opus 4.5 ↓1 0.00
- 6
DeepSeek-V3.2 0.00
rlm_benchmark
6 models we track have a score here.
- 1 Claude Sonnet 4.5 ↑1 0.67
- 2
Gemini 3 Pro ↑3 0.67
- 3 Claude Opus 4.5 ↑1 0.60
- 4
Gemini 3 Flash ↓3 0.47
- 5
Qwen3-Next-80B-A3B ↓2 0.37
- 6
DeepSeek-V3.2 0.33
sentinel2_urban_change_detection
6 models we track have a score here.
- 1
Qwen3-Next-80B-A3B ↑2 100.0
- 2 Claude Sonnet 4.5 100.0
- 3
Gemini 3 Pro ↑2 100.0
- 4 Claude Opus 4.5 100.0
- 5
DeepSeek-V3.2 ↑1 100.0
- 6
Gemini 3 Flash ↓5 100.0
tomato_virus_diagnosis_and_treatment
5 models we track have a score here.
- 1
Qwen3-Next-80B-A3B ↑2 0.00
- 2
Gemini 3 Pro ↑3 0.00
- 3 Claude Opus 4.5 ↑1 0.00
- 4
DeepSeek-V3.2 ↑2 0.00
- 5
Gemini 3 Flash ↓4 0.00
evaluate_single_model_accuracy
5 models we track have a score here.
- 1
Gemini 3 Pro ↑4 0.37
- 2
Gemini 3 Flash ↓1 0.31
- 3 Claude Opus 4.5 ↑1 0.22
- 4
Qwen3-Next-80B-A3B ↓1 0.04
- 5
DeepSeek-V3.2 ↑1 0.01
UCCB
6 models we track have a score here.
- 1
Gemini 3 Flash 4.69
- 2
DeepSeek-V3.2 ↑4 4.04
- 3 Claude Opus 4.5 ↑1 3.97
- 4 Claude Sonnet 4.5 ↓2 3.84
- 5
Gemini 3 Pro 3.20
- 6
Qwen3-Next-80B-A3B ↓3 3.17
Beat Gemini-2 in Poker
5 models we track have a score here.
- 1
Gemini 3 Pro ↑4 100.0
- 2 Claude Opus 4.5 ↑2 100.0
- 3
DeepSeek-V3.2 ↑3 100.0
- 4
Gemini 3 Flash ↓3 100.0
- 5
Qwen3-Next-80B-A3B ↓2 0.00
TraceQuest City Detective Benchmark
6 models we track have a score here.
- 1
Qwen3-Next-80B-A3B ↑2 0.18
- 2 Claude Sonnet 4.5 0.12
- 3 Claude Opus 4.5 ↑1 0.09
- 4
Gemini 3 Flash ↓3 0.07
- 5
DeepSeek-V3.2 ↑1 0.05
- 6
Gemini 3 Pro ↓1 0.00
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official Kaggle's Task Tuesday Benchmark 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 Kaggle's Task Tuesday Benchmark 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 Knowledge
- MMLU-Pro In use Harder 10-option version of MMLU with reasoning-focused questions across 14 subjects.
Intern-S2-397B 89.8%
- MMLU In use 57-subject multiple-choice exam questions from elementary to professional level. Scored by accuracy.
Seed1.8 92.3%
- SuperGPQA In use 26,529 graduate-level multiple-choice questions across 285 disciplines. Accuracy.
Qwen 3.6 Max (Preview) 73.9%
- WideSearch In use 200 bilingual broad info-seeking tasks: agent collects many atomic facts into a table. Scored by success rate and row/item F1.
Tencent Hy4 preview 83.9%
- BBEH micro average In use Models solve harder replacements for each of the 23 BIG-Bench Hard tasks, 4,520 examples in total. The score is the share of all examples answered correctly.
Gemma 4 31B IT 74.4%
- HLE-Diamond In use Models answer 1,000 cleaned expert questions from Humanity's Last Exam, 500 on reasoning and 500 on knowledge, closed book. Graded on exact answers. Claude Fable 5.2 77.9%
Data audit
8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 11 months ago, in December 2025, for Gemini 3 Flash. 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 (6 entries).
- Pass Models scored 6 models we track.
- Warning Newest score The newest came 11 months ago, in December 2025, for Gemini 3 Flash.
- Note Frontier models None of the top 10 has a score. The strongest tested is Gemini 3 Pro (#56).
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on kaggle.com: 6 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.