TabPFN-3.5-Thinking spends extra compute at fit time to get better at the metric you care about. In our new cookbook, that's worth €28,900 more in campaign value than XGBoost.
The setup: a bank ranks 200,000 customers by how likely they are to make a transaction, and only list. Whose list finds more of the right people?
→ Thinking ranks the right customers higher than XGBoost and standard TabPFN-3.5
→ In its top 1%, more than 95 in 100 make a transaction
→ Contacting the top 10% earns about 8% more than with XGBoost
Simulated figures (€100 per transaction, €2 per contact), so plug in your own numbers.
Try it out: docs.priorlabs.ai/cookbook/predi…
The setup: a bank ranks 200,000 customers by how likely they are to make a transaction, and only list. Whose list finds more of the right people?
→ Thinking ranks the right customers higher than XGBoost and standard TabPFN-3.5
→ In its top 1%, more than 95 in 100 make a transaction
→ Contacting the top 10% earns about 8% more than with XGBoost
Simulated figures (€100 per transaction, €2 per contact), so plug in your own numbers.
Try it out: docs.priorlabs.ai/cookbook/predi…
1
3
12
3,249
New TabPFN-3.5-Thinking is live on our API.
The wait is over: run it on your own data today, in the playground or via the API.
Built for high-stakes predictions where greater accuracy is worth the investment:
→ 95% win rate against the previous #1 model on large, real-world y to fitting time trade-off, with 32% less prediction time than before
→ Learns from up to 1M rows, 5x more than before
→ Best on grouped and temporal data, where patterns shift per customer, product and season
From power price forecasting to demand planning, customers are already putting it to work:
"𝘛𝘢𝘣𝘗𝘍𝘕 𝘛𝘩𝘪𝘯𝘬𝘪𝘯𝘨 𝘥𝘦𝘭𝘪𝘷𝘦𝘳𝘴 𝘣𝘦𝘵𝘵𝘦𝘳 𝘤𝘢𝘭𝘪𝘣𝘳𝘢𝘵𝘦𝘥 𝘲𝘶𝘢𝘯𝘵𝘪𝘭𝘦 𝘧𝘰𝘳𝘦𝘤𝘢𝘴𝘵𝘴 𝘵𝘩𝘢𝘯 𝘢𝘯𝘺 𝘮𝘦𝘵𝘩𝘰𝘥 𝘸𝘦 𝘩𝘢𝘷𝘦 𝘶𝘴𝘦𝘥, 𝘢𝘯𝘥 𝘪𝘵 𝘪𝘴 𝘯𝘰𝘸 𝘢 𝘬𝘦𝘺 𝘱𝘢𝘳𝘵 𝘰𝘧 𝘩𝘰𝘸 𝘸𝘦 𝘵𝘳𝘢𝘥𝘦."
- Christian Schütte, Head of Algo Trading and Quant, BD Energy
The wait is over: run it on your own data today, in the playground or via the API.
Built for high-stakes predictions where greater accuracy is worth the investment:
→ 95% win rate against the previous #1 model on large, real-world y to fitting time trade-off, with 32% less prediction time than before
→ Learns from up to 1M rows, 5x more than before
→ Best on grouped and temporal data, where patterns shift per customer, product and season
From power price forecasting to demand planning, customers are already putting it to work:
"𝘛𝘢𝘣𝘗𝘍𝘕 𝘛𝘩𝘪𝘯𝘬𝘪𝘯𝘨 𝘥𝘦𝘭𝘪𝘷𝘦𝘳𝘴 𝘣𝘦𝘵𝘵𝘦𝘳 𝘤𝘢𝘭𝘪𝘣𝘳𝘢𝘵𝘦𝘥 𝘲𝘶𝘢𝘯𝘵𝘪𝘭𝘦 𝘧𝘰𝘳𝘦𝘤𝘢𝘴𝘵𝘴 𝘵𝘩𝘢𝘯 𝘢𝘯𝘺 𝘮𝘦𝘵𝘩𝘰𝘥 𝘸𝘦 𝘩𝘢𝘷𝘦 𝘶𝘴𝘦𝘥, 𝘢𝘯𝘥 𝘪𝘵 𝘪𝘴 𝘯𝘰𝘸 𝘢 𝘬𝘦𝘺 𝘱𝘢𝘳𝘵 𝘰𝘧 𝘩𝘰𝘸 𝘸𝘦 𝘵𝘳𝘢𝘥𝘦."
- Christian Schütte, Head of Algo Trading and Quant, BD Energy
2
0
3
404
Try it in the playground: platform.priorlabs.ai/playground?tou…
Full results in the blog: priorlabs.ai/blog-posts/tab…
Full results in the blog: priorlabs.ai/blog-posts/tab…
0
0
0
80
📣 Less than a week to go for our TabPFN-Rel live session
If you're interested in relational machine learning and how we're bringing tabular foundation models like TabPFN-3.5 to the field, this one is not to miss.
We'll cover everything from RelArena (the new benchmark for N-Rel, our new harness for TFMs, brining them relational ML capabilities. Expect hands on examples on how to use the harness, and more news on what's next for us at Prior Labs!
📅 November 4th, 6PM CET
🔗 Register: app.livestorm.co/prior-labs/mac…
If you're interested in relational machine learning and how we're bringing tabular foundation models like TabPFN-3.5 to the field, this one is not to miss.
We'll cover everything from RelArena (the new benchmark for N-Rel, our new harness for TFMs, brining them relational ML capabilities. Expect hands on examples on how to use the harness, and more news on what's next for us at Prior Labs!
📅 November 4th, 6PM CET
🔗 Register: app.livestorm.co/prior-labs/mac…
0
1
5
750
Thinking now trains on up to 1M rows, 5x more than before, and is strongest on grouped and temporal data: patterns that shift per customer, product and season.
1
0
1
50
Full details in the blog. Link in comments.
Read the blog: priorlabs.ai/blog-posts/tab…
Sign up for early access and extra API credits: priorlabs.ai/blog-posts/tab…
Read the blog: priorlabs.ai/blog-posts/tab…
Sign up for early access and extra API credits: priorlabs.ai/blog-posts/tab…
0
0
0
52
How do you benchmark TabPFN against your current model?
New episode in our TabPFN Academy series: Philipp Singer (tag) walks through how he benchmarks TabPFN-3.5 against XGBoost.
Benchmarking is key if you want to evaluate TabPFN on your current baseline, so the tutorial and the notebook are yours: run it on your own data and see it for yourself.
youtube.com/watch?v=A7_rBO…
New episode in our TabPFN Academy series: Philipp Singer (tag) walks through how he benchmarks TabPFN-3.5 against XGBoost.
Benchmarking is key if you want to evaluate TabPFN on your current baseline, so the tutorial and the notebook are yours: run it on your own data and see it for yourself.
youtube.com/watch?v=A7_rBO…
1
0
6
626
🌌 Yes, we already told you about our agentic night in New York.
No, we’re not over it. More nights like this are coming, stay tuned!
No, we’re not over it. More nights like this are coming, stay tuned!
0
0
5
516
TabPFN-Rel gives your agent a relational prediction engine to call instead of a pipeline to write. The agent uses your business context and schema to define the task: what to predict, for which entities and over what period. TabPFN-Rel follows the relationships between tables, predicts with TabPFN-3.5.
It has guardrails built in for the mistakes agents typically make, like temporal leakage and incorrect handling of data types.
To get started, install the tabpfn-rel:relational-prediction skill in Claude Code or Codex and point your agent at your data.
More in the blog: priorlabs.ai/blog-posts/loo…
It has guardrails built in for the mistakes agents typically make, like temporal leakage and incorrect handling of data types.
To get started, install the tabpfn-rel:relational-prediction skill in Claude Code or Codex and point your agent at your data.
More in the blog: priorlabs.ai/blog-posts/loo…
0
0
3
595
📣 Live session for TabPFN-Rel incoming
Most business data is spread across many linked tables, but most ML tools expect one flat table. A few months ago, we released RelArena to push relational ML research. Now, join us to see how TabPFN-Rel makes predictions directly on ithout hand-built joins or feature engineering.
We'll cover:
→ RelArena - the new benchmark for relational ML and why we need it
→ What relational machine learning is, and why flattening tables by hand slows you down
→ How TabPFN-Rel works, and how it prevents data leakage
→ Defining prediction tasks in YAML and SQL, or letting your coding agent do it
→ Hands-on: building a seller churn model on a real e-commerce database in Colab
→ Live Q&A
📅 November 4th, 6PM CET
🔗 Register: app.livestorm.co/prior-labs/mac…
Most business data is spread across many linked tables, but most ML tools expect one flat table. A few months ago, we released RelArena to push relational ML research. Now, join us to see how TabPFN-Rel makes predictions directly on ithout hand-built joins or feature engineering.
We'll cover:
→ RelArena - the new benchmark for relational ML and why we need it
→ What relational machine learning is, and why flattening tables by hand slows you down
→ How TabPFN-Rel works, and how it prevents data leakage
→ Defining prediction tasks in YAML and SQL, or letting your coding agent do it
→ Hands-on: building a seller churn model on a real e-commerce database in Colab
→ Live Q&A
📅 November 4th, 6PM CET
🔗 Register: app.livestorm.co/prior-labs/mac…
1
0
10
554
"What will our sales be next quarter?"
For data teams, answering that usually means pulling data from a dozen tables, then joining, cleaning, engineering features and checking for leakage. Even with a coding agent, you're steering every step.
TabPFN-Rel is our harness for ional databases. Define your tables and prediction task in YAML and SQL, or let your coding agent do it. TabPFN-Rel handles the joins, feature engineering and prediction, and follows best practices in relational machine learning.
New in this release:
→ Even better performance: #1 using TabPFN-3.5-Plus and #2 using vanilla on RelArena-alpha, gaining up to 190 Elo compared to our initial release.
→ Its own repo and package: pip install tabpfn-rel
→ Over 100x faster preprocessing, making TabPFN-Rel five to 30 times faster than the next best model
→ New and improved skills ready to install to work even better with your coding agents (available as Claude Code and Codex plugins)
⭐️ Give us a star: github.com/PriorLabs/tabp…
For data teams, answering that usually means pulling data from a dozen tables, then joining, cleaning, engineering features and checking for leakage. Even with a coding agent, you're steering every step.
TabPFN-Rel is our harness for ional databases. Define your tables and prediction task in YAML and SQL, or let your coding agent do it. TabPFN-Rel handles the joins, feature engineering and prediction, and follows best practices in relational machine learning.
New in this release:
→ Even better performance: #1 using TabPFN-3.5-Plus and #2 using vanilla on RelArena-alpha, gaining up to 190 Elo compared to our initial release.
→ Its own repo and package: pip install tabpfn-rel
→ Over 100x faster preprocessing, making TabPFN-Rel five to 30 times faster than the next best model
→ New and improved skills ready to install to work even better with your coding agents (available as Claude Code and Codex plugins)
⭐️ Give us a star: github.com/PriorLabs/tabp…
3
0
5
657
Also announcement blog with more info: priorlabs.ai/blog-posts/loo…
0
0
0
152
How much stock of a life-saving drug should a pharmaceutical company order today, when future demand is uncertain? Order too little and patients go without. Order too much and you pay to store expensive stock.
Rahul Joshi from tulanā wrote a guest blog on how they solve this customers. The demo uses synthetic data modeled on a real customer: 7 years of weekly demand (5 for history, 2 held out), sharp seasonal surges, missing weeks, and a 6-week order lead time.
Their approach:
1. Build a probabilistic demand forecast
2. Simulate many possible futures and pick the re-order policy with the lowest expected cost
Fill rate and total cost on the 2 held-out years:
· Empirical quantiles (no seasonality): 83.3%, $87M
· Linear quantile regression (smooths out surges): 97.4%, $43M
· Tuned Quantile Regression Forest: 97.2%, $41M
· TabPFN-TS (no imputation, feature engineering or tuning): 98.6%, $37M
TabPFN had the highest fill rate and the lowest cost, $4M below the next best method.
Full post with code: priorlabs.ai/blog-posts/how…
Rahul Joshi from tulanā wrote a guest blog on how they solve this customers. The demo uses synthetic data modeled on a real customer: 7 years of weekly demand (5 for history, 2 held out), sharp seasonal surges, missing weeks, and a 6-week order lead time.
Their approach:
1. Build a probabilistic demand forecast
2. Simulate many possible futures and pick the re-order policy with the lowest expected cost
Fill rate and total cost on the 2 held-out years:
· Empirical quantiles (no seasonality): 83.3%, $87M
· Linear quantile regression (smooths out surges): 97.4%, $43M
· Tuned Quantile Regression Forest: 97.2%, $41M
· TabPFN-TS (no imputation, feature engineering or tuning): 98.6%, $37M
TabPFN had the highest fill rate and the lowest cost, $4M below the next best method.
Full post with code: priorlabs.ai/blog-posts/how…
2
0
8
506
Prior Labs retweeted
The difference between Jev and TFMs boil down to zero-shot classification vs grounded predictions. But because the two share a lot of keywords in descriptions, we get a lot of questions about how the two are related! So I've been sneaking it into my talks over the last few weeks!
ications for light-weight and cheap decision routing and (possibly) low stakes classifications. How? unstructured in -> structure out.
𝐓𝐚𝐛𝐮𝐥𝐚𝐫 𝐟𝐨𝐮𝐧𝐝𝐚𝐭𝐢𝐨𝐧 𝐦𝐨𝐝𝐞𝐥𝐬 𝐥𝐢𝐤𝐞 𝐓𝐚𝐛𝐏𝐅𝐍: grounded predictions (classify labels and regress values) based on ✨ learned data ✨ over tables. How? structure (tabular data, and a lot of it) in -> grounded, explainable predictions out. Definitely not zero-shot
There's a world where these two work together for predictive agents
ications for light-weight and cheap decision routing and (possibly) low stakes classifications. How? unstructured in -> structure out.
𝐓𝐚𝐛𝐮𝐥𝐚𝐫 𝐟𝐨𝐮𝐧𝐝𝐚𝐭𝐢𝐨𝐧 𝐦𝐨𝐝𝐞𝐥𝐬 𝐥𝐢𝐤𝐞 𝐓𝐚𝐛𝐏𝐅𝐍: grounded predictions (classify labels and regress values) based on ✨ learned data ✨ over tables. How? structure (tabular data, and a lot of it) in -> grounded, explainable predictions out. Definitely not zero-shot
There's a world where these two work together for predictive agents
0
3
4
770
Prior Labs retweeted
As a long time Kaggler who is now working on Tabular Foundation Models at @prior_labs it is so cool to see how TabPFN is the hot topic in tabular competitions on Kaggle right now. Participants pushing for the highest performance using TabPFN in their solutions is a real testament s have advanced over the last few months. Expect for this to only improve more and more in the future!
2
9
71
3,089
Last night our Co-Founder and Co-CEO @FrankRHutter received the German AI Innovation Award (Deutscher KI-Preis) from @welt .
The prize recognizes Frank Hutter's research in machine learning, and TabPFN as a key milestone in that story.
It started in his lab at the University @noahholl (first author of TabPFN & Prior Labs Co-Founder & CTO) and Samuel Müller asked whether the approach behind large language models could work on tables. The answer was the first foundation model for tabular data, published in Nature in January 2025.
Frank accepted the award on behalf of everyone behind TabPFN: the Freiburg researchers who built the first versions, and the Prior Labs team who turned that research into a model companies now use to forecast demand, price credit risk, find patterns in supply chains to name a few.
It was an exceptional night for Freiburg. Frank is the second Freiburg-affiliated winner of this prize, after Wolfram Burgard in 2022.
Congratulations to the other winners. @BoschGlobal received the AI Application Award for scaling AI in hardware and industrial settings, and @DuniaInnovation won the AI Startup Award. We’re working with the Dunia team on a benchmark and are glad to see them recognized.
Thank you to WELT and the jury, we are highly proud to see Frank's work recognized.
The prize recognizes Frank Hutter's research in machine learning, and TabPFN as a key milestone in that story.
It started in his lab at the University @noahholl (first author of TabPFN & Prior Labs Co-Founder & CTO) and Samuel Müller asked whether the approach behind large language models could work on tables. The answer was the first foundation model for tabular data, published in Nature in January 2025.
Frank accepted the award on behalf of everyone behind TabPFN: the Freiburg researchers who built the first versions, and the Prior Labs team who turned that research into a model companies now use to forecast demand, price credit risk, find patterns in supply chains to name a few.
It was an exceptional night for Freiburg. Frank is the second Freiburg-affiliated winner of this prize, after Wolfram Burgard in 2022.
Congratulations to the other winners. @BoschGlobal received the AI Application Award for scaling AI in hardware and industrial settings, and @DuniaInnovation won the AI Startup Award. We’re working with the Dunia team on a benchmark and are glad to see them recognized.
Thank you to WELT and the jury, we are highly proud to see Frank's work recognized.
1
2
14
834