iiinova — where your digital self keeps evolving

Kaggle @kaggle.com · 10/06/2026

1H-VideoQA is now available on Kaggle Benchmarks! Developed by Google DeepMind back in 2024 (Antoine Yang) and now updated with latest SOTA models, 1H-VideoQA is a 101-prompt benchmark for long-context video comprehension and temporal episodic reasoning across hour-long YouTube footage.

130

Kaggle @kaggle.com · 02/06/2026

Get ready - tomorrow June 3 at 9:30 AM PT / 12:30 PM ET, the 5-Day AI Agents: Intensive Vibe Coding team is going live on YouTube with Google Cloud!

Promotional graphic for a Google Cloud live stream titled "SNEAK PEEK Vibe coding AI agents course with Kaggle." The image features headshots of three speakers with their names below them: Anant Nawalgaria, Brenda Flynn and Smitha Kolan.

220

Kaggle @kaggle.com · 01/06/2026

We just opened the collective ML expertise of Kaggle’s community – discussions, solution writeups, debugging threads – to your coding agents. The new Kaggle CLI (v2.2.0) makes this knowledge your agent’s knowledge.

Terminal window titled "Kaggle CLI v2.2.0" showing the command "kaggle forums topics show 702989" and its output: a discussion topic titled "Kaggle CLI Update: Forums, Competition Topics, Benchmarks, and OAuth! (v2.2.0)" by Steve Messick, posted 2026-05-27, with 25 votes and 2 comments, followed by the beginning of the post body announcing the release.

220

Kaggle @kaggle.com · 21/04/2026

Registration is now open for the 5-Day AI Agents: Intensive Vibecoding Course with Google 🚀 This no-cost course is designed to help builders learn how to design, build, and use AI agents using the latest concepts, technologies and skills.

Promotional graphic for a "5-Day AI Agents Intensive Vibecoding Course with Google" by Kaggle and Google, scheduled for June 15 - 19, 2026. The image features playful illustrations of a laptop, coding icons, and abstract geometric shapes.

152

Kaggle @kaggle.com · 27/03/2026

Two weeks into the Measuring Progress Toward AGI - Cognitive Abilities hackathon, the benchmarks being built by the Kaggle community are already incredible. The Kaggle team Nick Kango and authors of the paper the hackathon is based on Dr Ryan Burnell, Oran Kelly are going LIVE to talk with you.

121

Kaggle @kaggle.com · 19/03/2026

We’re opening up the Kaggle toolbox to everyone. 🛠️ Today, we’re launching Community Hackathons - a free, self-serve way for you to host your own AI challenges. Whether you're an educator, a meetup lead, or just have a big idea, you can now build, judge and award prizes (up to $10k!).

110

Kaggle @kaggle.com · 17/03/2026

Earlier today, Google DeepMind released a new paper proposing a scientific framework for measuring the cognitive abilities of AI systems on the path to AGI. To better measure these capabilities, we’re partnering with them to launch a hackathon - Measuring Progress Toward AGI: Cognitive Abilities.

192

Kaggle @kaggle.com · 12/03/2026

📢 Exciting News! You can now receive notifications for Benchmarks on Kaggle! 🔔 You can now follow a benchmark to stay updated with alerts for new benchmark versions, new models added on leaderboards, and notifications for benchmark owners when new models are available to run.

130

Kaggle @kaggle.com · 19/02/2026

Four-in-a-Row is a “solved” game. Frontier LLMs still can’t play it reliably. 📢 We just launched a new Game Arena leaderboard to test how models reason step-by-step, maintain a mental board and plan moves - no minimax, no game-tree shortcuts.

A Kaggle "Game Arena Four in a Row Leaderboard" comparing 10 AI models. The table ranks models by (Internal) Elo, Average Output Tokens, and Average Inference Cost.

Top Performer: Gemini 3 Pro Preview (477 Elo, 11.90¢ per turn).

Runner Up: GPT-5.2 (450 Elo, 14.27¢ per turn).

Mid-Range: o3 (313 Elo), Grok 4 (313 Elo), and Gemini 3 Flash Preview (312 Elo).

Efficiency Leader: DeepSeek V3.2 ranks 10th (0 Elo) but features the lowest cost at 0.33¢ per turn.

The footer notes that ratings use the Bradley-Terry algorithm based on 80 games per model pair.

171

Kaggle @kaggle.com · 12/02/2026

📢 Exciting News! We are transitioning the Kaggle CLI and the `kagglehub` Python library out of “beta” and into a stable, production-ready state. As part of this release, we’re introducing several new features like support for multiple API tokens and more!

160

Kaggle @kaggle.com · 04/02/2026

What a show! 🏆 A huge thank you to everyone who tuned in and to our amazing partners @gmhikaru.bsky.social Nick Schulman, Liv Boeree, @dougpolkvids for the fantastic commentary and analysis across all three games, Poker, Chess and Werewolf.

Graphic for the "AI Poker Showdown Final Result," hosted by Kaggle. The image displays the completed tournament bracket showing o3 as the overall winner after defeating GPT 5.2 in the Day 3 Championship.

The final bracket highlights:
Day 1 Winners: o3, Gemini 3 Flash, GPT 5.2, and Opus 4.5.
Day 2 Winners: o3 (defeating Gemini 3 Flash) and GPT 5.2 (defeating Opus 4.5).
Final Result: o3 is crowned the champion

051

Kaggle @kaggle.com · 04/02/2026

📢The Grand Finale is here! 🏆 What happens when a chess Grandmaster and a Poker legend analyze AI? ♟️🃏

Graphic for "Game Arena AI Poker Showdown Day 3," a live event hosted by Kaggle and Google DeepMind on Wednesday, Feb 4th, from 9:30–11:30 AM PT. The image features a playful illustration of a robot at a poker table with a lightbulb over its head, showcasing the championship matchup: GPT 5.2 vs o3

130

Kaggle @kaggle.com · 03/02/2026

That’s a wrap on the semi-finals of the Game Arena! We have our Poker and Chess finalists locked in, and in Werewolf, the detective levels are off the charts.

Graphic for the "AI Poker Showdown," a tournament bracket hosted by Kaggle. The image displays the results of a multi-day competition leading to a Day 3 Championship match between o3 and GPT 5.2.

The bracket illustrates the following matchups: Day 1: o3 vs. Deepseek 3.2, Grok 4 vs. Gemini 3 Flash, GPT 5.2 vs. Gemini 3 Pro, and Opus 4.5 vs. Sonnet 4.5.
Day 2: o3 vs. Gemini 3 Flash and GPT 5.2 vs. Opus 4.5.
Day 3: The final showdown between o3 and GPT 5.2.

151

Kaggle @kaggle.com · 03/02/2026

It's the semi-finals today! Four models remain, and the stakes are doubling. 🃏♟️We’re live for the Poker Semi-Finals, Chess deep dives, and the penultimate Werewolf rounds!

Graphic for "Game Arena AI Poker Showdown," a live event hosted by Kaggle and Google DeepMind on Tuesday, Feb 3rd, from 9:30–11:30 AM PT. The image features a playful illustration of a robot at a poker table with a lightbulb over its head, surrounded by matchups for the tournament: GPT 5.2 vs Opus 4.5 and Gemini 3 Flash vs o3

131

Kaggle @kaggle.com · 02/02/2026

Day 1 of Game Arena is officially in the books! Congratulations to our AI poker showdown semi-finalists o3, Gemini 3 Flash, GPT 5.2, and Opus 4.5!

281

Kaggle @kaggle.com · 02/02/2026

🎬 We’re live! Watch GMHikaru and Nick Schulman break down the first round of the poker bracket and chess newcomer matches.

Game Arena AI Poker Showdown," a live event hosted by Kaggle and Google DeepMind on Monday, Feb 2nd, from 9:30–11:30 AM PT. The image features a playful illustration of a robot at a poker table with a lightbulb over its head, surrounded by matchups for the tournament: o3 vs. DeepSeek 3.2, Grok 4 vs. Gemini 3 Flash, GPT 5.2 vs. Gemini 3 Pro, and Opus 4.5 vs. Sonnet 4.5.

151

Kaggle @kaggle.com · 02/02/2026

Game Arena kicks off today! 📣 Top AI models compete in Poker, Werewolf, and Chess, testing reasoning, social strategy, and risk management. 🎙️ Co-hosted by GM Hikaru & Poker Hall-of-Famer Nick Schulman: www.youtube.com/GMHikaru 🗓️ Feb 2–4 | 9:30–11:30 AM PT More info 👇

1215

Kaggle @kaggle.com · 29/01/2026

📌 Mark Your Calendar: Live Game Arena Event This Monday! We are releasing two new games, Poker and Werewolf, along with an updated Chess leaderboard next Monday, February 2, running daily from 9:30 AM PT to 11:30 AM PT through February 4

2155

Kaggle @kaggle.com · 14/01/2026

🚀 Introducing Community Benchmarks on Kaggle! As AI evolves at an unprecedented pace, measuring intelligence requires more than a few AI research labs alone – it requires the imagination and collective expertise of the global community. That’s why we’re launching Community Benchmarks.

120

Kaggle @kaggle.com · 18/12/2025

🏆 Announcing the winners of the Agents Intensive Capstone Project! 🎉 We're excited to announce the top 12 teams who showcased exceptional creativity & technical skill using AI agents! Check out their innovative projects & learn more about their submissions here: www.kaggle.com/competitions...

020

Kaggle @kaggle.com · 11/12/2025

🚀 New on Kaggle Benchmarks: DeepSearchQA developed by Google DeepMind! This benchmark focuses on complex web research tasks and tests agent comprehensiveness. Check the leaderboard: www.kaggle.com/benchmarks/g...

A screenshot of the Kaggle DeepSearchQA leaderboard, showing the top five ranked models.

110

Kaggle @kaggle.com · 11/12/2025

📢 The FACTS Benchmark Suite is now live on Kaggle! Developed by Google DeepMind and Google Research, this suite measures LLM factuality across four dimensions: Parametric knowledge, Search, Multimodal understanding & Grounding. Explore the leaderboard: www.kaggle.com/benchmarks/g...

A screenshot of the Kaggle FACTS Benchmark Suite leaderboard. The table displays several large language models like GPT-4, Gemini, and others, ranked by their overall FACTS Score and performance breakdown in the four categories: Parametric Knowledge, Search, Multimodal Understanding, and Grounding. The overall score and dimension scores are visible.

030

Kaggle @kaggle.com · 09/12/2025

🚀 Benchmark your AI across India’s languages with IndicGenBench! Developed by Google DeepMind, this benchmark spans 29 Indic languages, including first-ever evaluation data for 18 Indic languages. It supports language tasks like summarization, translation and question answering.

A screenshot of the IndicGenBench leaderboard on Kaggle Benchmarks. The leaderboard ranks various AI models based on their performance across 29 Indic languages on generative tasks. The top models and their scores are visible, showing a comparison of AI performance on tasks like cross-lingual summarization, machine translation and question answering for Indian languages.

123

Kaggle @kaggle.com · 03/12/2025

🚀 Feature Update on Kaggle Benchmark You can now download Kaggle Benchmark leaderboard results! Compare your favorite models with a simple CURL command or download the full CSV directly for deeper analysis. Get started: www.kaggle.com/benchmarks

030

Kaggle @kaggle.com · 22/10/2025

♟️ Expanding Game Arena: Introducing Chess Openings A new benchmark that tests reasoning beyond memorization. Each game starts from one of 20 popular openings, pushing models to adapt and think strategically rather than rely on learned patterns.

Image of the Kaggle Game Arena Chess Openings leaderboard. The leaderboard table displays the top AI models ranked by performance, with columns showing model name, score, and number of games played. The image highlights the competitive rankings and emphasizes the new Chess Openings benchmark, which starts each game from one of 20 popular two-ply openings to test adaptability and strategic reasoning beyond memorization.

141

Kaggle @kaggle.com · 17/10/2025

Check out the leaderboard here: 👇 www.kaggle.com/benchmarks/c...

Leaderboard showing Gemini-2.5-Flash achieving the top score of 94.8% on the Global MMLU Lite multilingual benchmark. This evaluation dataset tests models across 16 languages for cultural and linguistic biases.

000

Kaggle @kaggle.com · 10/09/2025

🚀 New Benchmark Launch: SimpleQA Verified! We’ve partnered with Google DeepMind and Google Research to launch a curated 1,000-prompt benchmark designed to provide a more reliable and challenging evaluation of LLM short-form factuality. Check out the leaderboard here: www.kaggle.com/benchmarks/d...

Screenshot of a new benchmark - SimpleQA Verified launched in partnership with Google DeepMind and Google Research.

021

Kaggle @kaggle.com · 22/08/2025

🏆 Results are in! In the first #KaggleGameArena — Chess Text Input — AI models faced off using only text inputs (no tools, no move validation) in 40+ matches per pairing to build a robust Elo-like ranking ♟️ www.kaggle.com/benchmarks/k...

A screenshot of the chess text input benchmark leaderboard.

132

Kaggle @kaggle.com · 07/08/2025

What a show! The Kaggle Game Arena AI Chess Tournament is complete — and O3 takes the win! 🏆 Big thanks to @magnuscarlseny.bsky.social , @gmhikaru.bsky.social, @gothamchess.bsky.social and GM David Howell for the fantastic commentary and analysis on Chessom and TakeTakeTakeApp.

063

Kaggle @kaggle.com · 05/08/2025

What an exciting start to the Kaggle Game Arena AI chess exhibition tournament ♟️! The first round is complete, and we have our four semi-finalists! Congratulations to o4-mini, o3, Gemini 2.5 Pro & Grok 4! Come back tomorrow! Semi-finals kick off, August 6th, at 10:30 am PT.

143

Kaggle @kaggle.com · 05/08/2025

Let the games begin! It's time to watch eight LLMs compete in the first round of head-to-head match-ups. #KaggleGameArena

173

Kaggle @kaggle.com · 05/08/2025

It’s Day 1 of the Kaggle Game Arena AI chess exhibition tournament ♟️! Tune in today at 10:30AM PT to watch 4 head-to-head AI matchups 🤖 in a single-elimination bracket

100

Kaggle @kaggle.com · 05/08/2025

Think you can predict the winner of our AI chess exhibition tournament? 🧠🏆 Reply to this post with your filled-out bracket to let us know who you think will take home the gold medal!

010

Kaggle @kaggle.com · 04/08/2025

The inaugural #KaggleGameArena AI chess exhibition tournament kicks off live tomorrow. For the next 3 days, August 5-7, tune in daily at 10:30 am PST, and catch commentary from @gmhikaru.bsky.social, @gothamchess.bsky.social and @magnuscarlseny.bsky.social ⬇️

151

Kaggle @kaggle.com · 04/08/2025

Don’t forget to check out the Kaggle AI chess exhibition tournament tomorrow. The matches take place daily from August 5-7, with streams starting at 10:30 AM PT on kaggle.com/game-arena.

152

Kaggle @kaggle.com · 04/08/2025

Games are also a fantastic proxy for a wide range of real-world skills. They test a model's ability in strategic planning, reasoning, memory, adaptation, and even "theory of mind" – understanding an opponent's thoughts.

110

Kaggle @kaggle.com · 04/08/2025

📢 Introducing Kaggle Game Arena: a new, open benchmark platform where top AI models compete in complex, strategic games in streamed match-ups. We're charting new frontiers for trustworthy AI evaluation and it begins with chess — a classic proving ground for system intelligence.

1124

Kaggle @kaggle.com · 15/07/2025

✔️ Join the waitlist for early access to Kaggle Benchmarks! Kaggle Benchmarks is the fastest, easiest way to test new models. Let Kaggle handle infrastructure while you focus on AI breakthroughs and benefit from competition-grade rigor. Sign up here: goo.gle/kaggle-benchmarks-waitlist

000

Kaggle @kaggle.com · 14/07/2025

📣 ICML 2025 Alert! Find Kaggle at Booth #121. Meet our team, explore an interactive demo, & our new community platform for building and sharing top models evaluations. ➕ learn more about Kaggle team's upcoming talk on GenAI evaluation! #ICML2025

000

Kaggle @kaggle.com · 13/03/2025

📢 You can now connect your Google Colab notebooks directly to Kaggle's Jupyter Servers! Access Kaggle's powerful compute resources like GPUs, TPUs & large datasets from your preferred editor, like Colab or VS Code. Try it now! 👇 www.kaggle.com/discussions/...

010

Kaggle @kaggle.com · 22/01/2025

Kaggle’s notebooks workspace is a no-cost, no set-up way to bring reproducible data science and ML projects to life. We’ll share features that help you get the most out of this resource. #WaysOfKaggling

120

Kaggle @kaggle.com · 21/01/2025

Competitions are a spectator sport, too! Even if you don’t compete, there’s a wealth of knowledge, reproducible code, and resources openly shared by the community. Browse code and solution write-ups shared by competitors: www.kaggle.com/competitions...

110

Kaggle @kaggle.com · 21/01/2025

Interested in applied problems? Did you know Kaggle also hosts Analytics competitions with more open-ended objectives? Right now you can explore NFL data in their annual Big Data Bowl or learn how to build and share language variants of Gemma. Learn more www.kaggle.com/competitions...

100

Kaggle @kaggle.com · 21/01/2025

Competitions are the best ways to learn and hone skills on real-world problems. But, getting started can be intimidating. The monthly Playground Tabular Series is the perfect, low-stakes way to begin. 🧑‍🏫 Check out the latest one here 🔗 www.kaggle.com/competitions...

A screenshot of the playground tabular series page on Kaggle.

100

Kaggle @kaggle.com · 21/01/2025

Today for #WaysToKaggle, we're diving into Kaggle Competitions! It’s where the community comes to learn, collaborate, and innovate together. Let’s explore! 🚀 www.kaggle.com/competitions

Today, we're diving into Kaggle Competitions! It’s where the community comes to learn, collaborate, and innovate together.

110

Kaggle @kaggle.com · 21/01/2025

Competitions are the best ways to learn and hone skills on real-world problems. But, getting started can be intimidating. The monthly Playground Tabular Series is the perfect, low-stakes way to begin. 🧑‍🏫

100

Kaggle @kaggle.com · 21/01/2025

Today for #WaysToKaggle, we're diving into Kaggle Competitions! It’s where the community comes to learn, collaborate, and innovate together. Let’s explore! 🚀 www.kaggle.com/competitions

100

Kaggle @kaggle.com · 17/01/2025

You can also browse top models by competition to see which ones are popular and working well for a specific benchmark task. Plus, explore publicly and fork shared code using the models you’re interested in. 🔗 www.kaggle.com/competitions...

100

Kaggle @kaggle.com · 17/01/2025

Kaggle Models is one of the newest parts of Kaggle’s platform. Discover how the community uses top models from publishers like Qwen, AI at Meta, Cohere and more to solve real-world tasks in competitions. #WaysToKaggle 🔗 www.kaggle.com/models

120

Kaggle @kaggle.com · 16/01/2025

Once you’re comfortable with Kaggle’s platform, join a Getting Started competition like the LLM Classification Finetuning based on data from LMArena. Check out the /code tab for helpful starter notebooks! 🔗 www.kaggle.com/competitions...

100