GPT-6 Astra and GPT-5.6 Sol are now powering HiPeople
GPT-6 Astra and GPT-5.6 Sol are now powering HiPeople
HiPeople now uses GPT-6 Astra and GPT-5.6 Sol for selected tasks across our platform.
For us, the exciting part about a new generation of AI models isn’t simply that they’re more powerful. It’s all about what that additional intelligence allows us to build in the context of running hiring.
An AI agent running hiring end to end has to solve very different problems along the way. Some require speed and focused execution. Others require deep reasoning across large amounts of hiring context.
That’s where the latest generation of models becomes particularly interesting.
Where stronger reasoning matters in hiring
Consider one of the hardest problems in hiring: making sense of everything you know about a candidate in the context of a specific job.
There can be dozens or even hundreds of relevant signals: the requirements of the role, a candidate’s experience and career progression, application and screening data, interview responses, assessments, recruiter feedback, hiring manager feedback, and everything else learned throughout the process.
Making sense of all of this isn’t primarily a generation problem. It’s a reasoning problem.
The model needs to understand the job while interpreting evidence from many different sources. It needs to distinguish strong signals from weak ones, connect information across the hiring journey, reconcile signals that may point in different directions, and arrive at a useful conclusion without losing the nuance underneath it.
This is exactly the kind of work where advances in frontier models can make a meaningful difference.
GPT-6 Astra and GPT-5.6 Sol
We’re now using GPT-6 Astra and GPT-5.6 Sol for selected tasks across HiPeople.
For particularly demanding reasoning tasks, these models give HiPeople additional capability to deeply synthesize hiring context and produce precise, useful outputs.
That can mean making deeper sense of a candidate against the requirements of a role, connecting information gathered at different stages of the hiring process, or reasoning across a large body of hiring data before producing a recommendation.
The result isn’t simply more powerful AI. It’s greater precision and even stronger hiring recommendations.
The right intelligence for each agent task
Not every task across the hiring journey requires the same kind of intelligence.
Extracting structured information, executing a focused workflow, evaluating a complex body of evidence, and synthesizing an entire hiring journey are fundamentally different problems.
That’s why we bring together the strengths of different models, matching each task with the intelligence best suited to deliver the strongest result (at a reasonable price).
As models continue to improve, this architecture lets the intelligence underneath HiPeople improve with them.
What this means for agentic hiring
Ultimately, the interesting part isn’t that HiPeople has two new models.
It’s what increasingly capable models allow an AI hiring agent to understand and do.
As reasoning improves, an agent can make deeper sense of information across the entire hiring journey rather than treating sourcing, screening, interviewing, and decision-making as isolated steps.
More context can be understood together. More complex relationships between signals can be identified, and the recommendations delivered to hiring teams can become increasingly precise.
GPT-6 Astra and GPT-5.6 Sol are the latest additions to the model stack helping us get there.
Make sure to subscribe to the HiPeople blog to stay up to date on new models, product releases, and what they mean for the future of hiring.
