> ## Documentation Index
> Fetch the complete documentation index at: https://docs.priorlabs.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# RelArena-α, KV cache in the API, performance improvements, and more
> Explore RelArena-α, TabPFN-Rel, and RPI, plus KV-cached inference through the API, decoder attention readout, calibrated regression, and faster local fitting.
This update covers everything we've shipped since May 2026: `tabpfn` 8.4.0, `tabpfn-client` 0.4.2, the latest shipped TabPFN API release, and `tabpfn-extensions` 0.6.0. During this time, we've also introduced [TabPFN Cookbooks](/cookbook) and [Research Applications](/tabpfn-in-action/research), sharing how TabPFN can be applied for specific use cases. Check them out—and contribute your own.
## Open-source package
`tabpfn` 8.4.0 ships with improvements across regression, fitting performance, and Apple Silicon inference:
* **Better uncertainty estimates for regression.** Made for forecasting, risk estimation, and other workflows where interval quality matters alongside point predictions. `TabPFNRegressor` can now calibrate its ensemble distribution against held-out data using NLL or CRPS, improving probabilistic outputs and predicted quantiles ([#1172](https://github.com/PriorLabs/TabPFN/pull/1172)).
* **Faster, lower-memory fitting on large datasets.** Made for large or wide datasets—we've cut transient memory use and fit time by roughly two-thirds in benchmarks on a 5.3 GB numeric table. The data-cleaning path now avoids redundant float64 copies, assembles encoded arrays in place, and performs pandas dtype casting more efficiently ([#1173](https://github.com/PriorLabs/TabPFN/pull/1173)).
* **Local inference on Apple Silicon.** When running TabPFN on a Mac, you can now use MPS inference with MLX-backed FlashAttention and autocast, along with fixes for out-of-memory recovery and incorrect linear-layer outputs.
We also shipped:
* `n_estimators` now defaults to `"auto"` ([#1171](https://github.com/PriorLabs/TabPFN/pull/1171)).
* Calibrated regression now fails explicitly for batched predictions instead of returning uncalibrated results ([#1172](https://github.com/PriorLabs/TabPFN/pull/1172)).
* DDP fine-tuning now distributes estimator activations across available GPUs. [Read the fine-tuning docs](/capabilities/fine-tuning#multi-gpu-fine-tuning-ddp).
* Regression now supports batched inference ([#1164](https://github.com/PriorLabs/TabPFN/pull/1164)).
### RelArena-α, TabPFN-Rel, and RPI
This week, we released [RelArena-α](https://github.com/PriorLabs/relarena) alongside a family of open-source tools for prediction tasks over relational data. The full benchmarking framework and accompanying tools are described in the [technical report](https://arxiv.org/abs/2608.16319).
[Read why we're focusing on relational](https://priorlabs.ai/blog-posts/introducing-relarena).
* **Compare relational models under a shared evaluation protocol and run benchmarked methods on your own relational data.** RelArena-α standardizes data loading, tuning, and evaluation across relational learning tasks, making results easier to reproduce and compare. The Relational Predictive Interface (RPI) provides a common interface for applying these methods to real-world prediction tasks.
* **Apply TabPFN to relational prediction tasks.** TabPFN-Rel is a relational harness for TabPFN-3 and TabPFN-3-Plus, and is currently the No. 1 model submission on RelArena-α. [Get started with the TabPFN-Rel cookbook](https://docs.priorlabs.ai/cookbook/relational_predictions_tabpfn_rel?utm_source=newsletter\&utm_campaign=relational).
## tabpfn-client and API
`tabpfn-client` 0.4.2 and the TabPFN API ship with improvements across repeated inference and integrations:
* **Up to 10× faster repeat inference with KV cache.** Best for large prediction volumes against the same training data and long-running interpretability workloads. `fit_with_cache` can now be requested during `fit()` through `tabpfn-client` 0.4.2 and the hosted TabPFN API, reusing the training-set computation so subsequent predictions avoid the full forward pass and can run in under a second. [Get started with KV cache](/capabilities/kv-cache).
We also shipped:
* TabPFN-3-Plus, including Thinking mode, can now run on Azure AI Foundry through a new estimator interface. [Get started with Azure AI Foundry](/integrations/foundry).
* `output_type="full"` now returns a consistent response type.
* Asynchronous fitting is now available through `tabpfn-client` 0.4.2 and the hosted TabPFN API, improving reliability. [See the API reference](/api-reference/getting-started).
* The MCP integration now supports asynchronous operations and the standard authentication path. [Read the MCP docs](/agentic/mcp).
## Extensions
`tabpfn-extensions` 0.6.0 ships with improvements across interpretability and client compatibility:
* **See which training rows drive a prediction.** Best for explaining borderline multiclass predictions, investigating unexpected outputs, and seeing how the training set affects results. The same readout can be viewed in target-conditioned embedding space or in the original feature space ([#1142](https://github.com/PriorLabs/TabPFN/pull/1142)).
We also shipped:
* The client-backed regressor now preserves `output_type="full"` and forwards requested output types correctly ([#354](https://github.com/PriorLabs/tabpfn-extensions/pull/354)).
## Migrating
Review these behavior changes, deprecations, and removals before upgrading production code.
### OSS (local / Hugging Face)
1. Review `n_estimators` when upgrading: it now defaults to `"auto"`, feature-coverage scaling applies only to `"auto"`, and an explicit integer is used exactly as provided. Replace `auto_scale_n_estimators`, which is deprecated and scheduled for removal in the next major release.
2. If you calibrate regression with `tuning_config` and need to score a batch of datasets, call `predict()` on each dataset individually: `predict_batched` does not support calibrated regression and now raises `NotImplementedError` instead of returning uncalibrated fused-batch predictions.
3. Use PyTorch 2.6.0 or newer for MPS inference.
### Extensions
1. Include every feature in the `dag` mapping passed to `TabPFNUnsupervisedModel.generate_synthetic_data()` or `impute()`; declare parentless features with an empty list.
### API
1. Reinstall the MCP integration to pick up the latest changes.