‹ All stories

flow-1

Model · 1 update · last updated Oct 2

Where it stands

As of Oct 2, 2026, Laminar's flow-1 is an RL-trained model that finds hard-to-spot failures in agent traces, matching GPT-6-sol detection quality at 23x lower cost on its benchmark.

Timeline · newest first
  1. Oct 2
    Official

    Laminar releases flow-1, an RL-trained model for finding failures in agent traces

    Laminar released flow-1, a model for understanding agent traces, trained with reinforcement learning to identify hard-to-spot failures and explain why they happened. On Laminar's benchmark, it matches GPT-6-sol in detection quality while being 23x cheaper, per the company. The goal is to make it affordable to investigate all production traces continuously instead of sampling. flow-1 runs inside the company's Signals agent, which treats the trace as a repo and each span as a file.

    Source: Laminar
How we label
Official
An institution announces something that has already happened, through its own channel or a founder or executive account.
Statement by …
A named person's views, predictions, plans or comments, including remarks reported by media when the original is unavailable.
Reported by …
Only media or another third party is saying it. We name the outlet.
Unconfirmed
Circulating online (posts, comments, leaks). No company, named person or outlet has confirmed it.

Items older than 72 hours never go in Top stories and show their original date.

DailyCitedRSSAboutPrivacy