A discrepancy between first- and third-party benchmark outcomes for OpenAI’s o3 AI mannequin is raising questions about the company’s transparency and mannequin testing practices.

When OpenAI unveiled o3 in December, the corporate claimed the mannequin may reply simply over a fourth of questions on FrontierMath, a difficult set of math issues. That rating blew the competitors away — the next-best mannequin managed to reply solely round 2% of FrontierMath issues accurately.

“As we speak, all choices on the market have lower than 2% [on FrontierMath],” Mark Chen, chief analysis officer at OpenAI, said during a livestream. “We’re seeing [internally], with o3 in aggressive test-time compute settings, we’re in a position to recover from 25%.”

Because it seems, that determine was possible an higher certain, achieved by a model of o3 with extra computing behind it than the mannequin OpenAI publicly launched final week.

Epoch AI, the analysis institute behind FrontierMath, launched outcomes of its impartial benchmark assessments of o3 on Friday. Epoch discovered that o3 scored round 10%, effectively beneath OpenAI’s highest claimed rating.

That doesn’t imply OpenAI lied, per se. The benchmark outcomes the corporate revealed in December present a lower-bound rating that matches the rating Epoch noticed. Epoch additionally famous its testing setup possible differs from OpenAI’s, and that it used an up to date launch of FrontierMath for its evaluations.

“The distinction between our outcomes and OpenAI’s is perhaps as a result of OpenAI evaluating with a extra highly effective inside scaffold, utilizing extra test-time [computing], or as a result of these outcomes had been run on a unique subset of FrontierMath (the 180 issues in frontiermath-2024-11-26 vs the 290 issues in frontiermath-2025-02-28-private),” wrote Epoch.

According to a post on X from the ARC Prize Basis, a company that examined a pre-release model of o3, the general public o3 mannequin “is a unique mannequin […] tuned for chat/product use,” corroborating Epoch’s report.

“All launched o3 compute tiers are smaller than the model we [benchmarked],” wrote ARC Prize. Typically talking, greater compute tiers might be anticipated to attain higher benchmark scores.

OpenAI’s personal Wenda Zhou, a member of the technical employees, said during a livestream last week that the o3 in manufacturing is “extra optimized for real-world use circumstances” and pace versus the model of o3 demoed in December. Consequently, it could exhibit benchmark “disparities,” he added.

“[W]e’ve finished [optimizations] to make the [model] extra value environment friendly [and] extra helpful on the whole,” Zhou mentioned. “We nonetheless hope that — we nonetheless suppose that — it is a a lot better mannequin […] You received’t have to attend as lengthy whenever you’re asking for a solution, which is an actual factor with these [types of] fashions.”

Granted, the truth that the general public launch of o3 falls wanting OpenAI’s testing guarantees is a little bit of a moot level, because the firm’s o3-mini-high and o4-mini fashions outperform o3 on FrontierMath, and OpenAI plans to debut a extra highly effective o3 variant, o3-pro, within the coming weeks.

It’s, nevertheless, one other reminder that AI benchmarks are finest not taken at face worth — notably when the supply is an organization with providers to promote.

Benchmarking “controversies” have gotten a typical incidence within the AI business as distributors race to seize headlines and mindshare with new fashions.

In January, Epoch was criticized for ready to reveal funding from OpenAI till after the corporate introduced o3. Many teachers who contributed to FrontierMath weren’t knowledgeable of OpenAI’s involvement till it was made public.

Extra lately, Elon Musk’s xAI was accused of publishing deceptive benchmark charts for its newest AI mannequin, Grok 3. Simply this month, Meta admitted to touting benchmark scores for a model of a model that differed from the one the company made available to developers.

Up to date 4:21 p.m. Pacific: Added feedback from Wenda Zhou, a member of the OpenAI technical employees, from a livestream final week.




Source link