“Google announced their big new state-of-the-art model yesterday, Gemini 3, on most measures better than any other model out there.”
Our reading, the sentence that was checked:
Gemini 3 is better than any other model on most measures.
Assessment: Supported Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, Hot take on Google’s Gemini 3, paragraph 1.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-hot-take-on-googles-gemini-3-53bbf6-a1
The wording above is the version this site observed on 2026-10-28. The source may have been edited after 2025-11-19. This page does not claim the wording was captured then.
What the claim commits to
- Kind
- Observation
- Deadline, as the speaker wrote it
- None stated. This site does not add one.
- Deadline state when assessed
- No horizon
- Deadline state on 2026-10-29
- No horizon (the release’s reference date)
Conditions
- on most measures (qualifier, stated by the speaker)
From the speaker
- Gemini 3
- announced yesterday
- on most measures
- better than any other model out there
Our interpretation
- Plain reading: at launch, Gemini 3 (Pro) scored first against the other frontier models on most of the widely cited benchmarks and leaderboards.
- 'Most measures' is read as a majority of the standard benchmarks reported at launch, including independent leaderboards, not every single test.
Counts as supported if: Launch-time benchmark results (company and independent) show Gemini 3 ranked first on a majority of standard measures against rival frontier models.
Counts as contradicted if: Launch-time results show rival models leading on most standard benchmarks, or Gemini 3 leading on only a minority.
Retrospective: these criteria were written after the outcome was public.
Current assessment
Was it true for the period described?
Outcome: Supported
The dated evidence matches what the claim committed to, under the criteria frozen before grading.
Scored on a plain reading. The speaker named no measure for this claim, so it was judged on the reading an informed reader would take, stated below under our interpretation. Both reviews had to find that reading fair.
- Evidence as of
- Deadline state when assessed
- No horizon
- Assessment recorded
- Policy version
- 2026-10-28.7
Rationale
At launch, Google reported that Gemini 3 Pro ranked first on LMArena and posted top scores on Humanity's Last Exam, GPQA Diamond and other benchmarks. A launch-day news summary of the comparison table says it beat Anthropic and OpenAI models in nearly all tests. The independent evaluator Artificial Analysis put it first overall on its Intelligence Index. There were exceptions: Claude edged it on SWE-bench Verified, and Artificial Analysis had it first on 5 of 10 component evals. His hedge 'on most measures' allows for such exceptions, so the claim holds on its plain reading.
Limitations
- Much of the comparison comes from Google's own chosen benchmarks.
- Independent component-level results show leadership on about half the evals, not an overwhelming majority.
Evidence (4 items)
Supports the claim: Google,
It tops the LMArena Leaderboard with a breakthrough score of 1501 Elo. It demonstrates PhD-level reasoning with top scores on Humanity’s Last Exam (37.5% without the usage of any tools) and GPQA Diamond (91.9%).
Why it matters here: Google reports that Gemini 3 Pro ranked first on LMArena and posted top scores on major reasoning benchmarks at launch.
- Publisher
- Kind of source
- Company disclosure: the company describing itself, not an independent finding
- Describes events of
- Public from
Supports the claim: Artificial Analysis,
The model outperforms all other models in Artificial Analysis Intelligence Index. It demonstrates strength across the board, coming in first in 5 of the 10 evaluations that make up Intelligence Index.
Why it matters here: An independent evaluator put Gemini 3 Pro first overall. It led 5 of 10 component evals, plus its own knowledge benchmark and MMMU-Pro, so it did not lead every measure.
- Where
- intro paragraph
- Publisher
- Artificial Analysis
- Kind of source
- Independent measurement
- Describes events of
- Public from
Supports the claim: OfficeChai,
Gemini 3 Pro, has posted dominant results across a comprehensive suite of industry benchmarks, outperforming competing models from Anthropic and OpenAI in nearly all tests.
Why it matters here: A launch-day news summary of the comparison table says Gemini 3 Pro beat Anthropic and OpenAI models on nearly all tests.
- Where
- paragraph 2
- Publisher
- OfficeChai
- Kind of source
- News report
- Describes events of
- Public from
Context only: OfficeChai,
76.2% on SWE-Bench Verified for single-attempt coding tasks, though Claude’s 77.2% edged it out on the latter.
Why it matters here: Shows an exception: Gemini 3 did not lead every measure. This fits his word 'most'.
- Publisher
- OfficeChai
- Kind of source
- News report
- Describes events of
- Public from
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, Hot take on Google’s Gemini 3
Google announced their big new state-of-the-art model yesterday, Gemini 3, on most measures better than any other model out there.
- Where
- paragraph 1
- Published
- 2025-11-19
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-hot-take-on-googles-gemini-3-53bbf6-a1: Current. Outcome: Supported. Evidence as of 2026-10-28. Public date not recorded.
Recorded events
Assessment publishedBy this site
First assessment published after both reviews and the final-text check passed.
About assessment marcus-hot-take-on-googles-gemini-3-53bbf6-a1.