Basis completes a tax workbook 2x faster with GPT-6 Astra

OpenAI Blog2026-09-28 08:000 views

GPT-6 Astra completed a 50-tab tax workbook twice as fast as GPT-5.6 Sol, and its stronger understanding of user intent gives Basis more confidence in real-world use.

Source：https://openai.com/index/basis-tax-workbook-with-astra

My favorites

PreviousAre you a Codex Original?NextYour Website Images Are Part of Search Now. Are You Treating Them That Way?

Hypergraph Generation via Structured Stochastic Diffusion

arXiv:2605.05024v2 Announce Type: replace-cross Abstract: Hypergraphs model higher-order interactions, but realistic hyp

0arXiv stat.CO

Sparse Probabilistic Richardson Extrapolation

arXiv:2604.02072v2 Announce Type: replace-cross Abstract: Almost every numerical task can be cast as extrapolation with

0arXiv stat.CO

Dynamic Bayesian Predictive Stacking via Markovian Spatiotemporal Propagation

arXiv:2602.08544v3 Announce Type: replace-cross Abstract: This manuscript develops computationally efficient online lear

0arXiv stat.CO

Coherent forecast combination for linearly constrained multiple time series

arXiv:2412.03429v2 Announce Type: replace-cross Abstract: Forecasting linearly constrained multiple time series, as thos

0arXiv stat.CO

New optimisations for exact, scalable 2D Gaussian process inference, with applications to astronomical time-series

arXiv:2611.09198v1 Announce Type: cross Abstract: Astronomical datasets often consist of a grid of datapoints in two or

0arXiv stat.CO

A tractable transition function for the Wright-Fisher diffusion with selection via gamma duality

arXiv:2611.08623v1 Announce Type: cross Abstract: The transition function of the Wright-Fisher diffusion with selection

0arXiv stat.CO

Basis completes a tax workbook 2x faster with GPT-6 Astra

Related posts

Hypergraph Generation via Structured Stochastic Diffusion

Sparse Probabilistic Richardson Extrapolation

Dynamic Bayesian Predictive Stacking via Markovian Spatiotemporal Propagation

Coherent forecast combination for linearly constrained multiple time series

New optimisations for exact, scalable 2D Gaussian process inference, with applications to astronomical time-series

A tractable transition function for the Wright-Fisher diffusion with selection via gamma duality