Capability gains

Google previews Gemini 1.5 Pro with one-million-token context

Google introduced Gemini 1.5 Pro in limited private preview, reporting performance comparable to Gemini 1.0 Ultra and experimental processing of up to one million multimodal tokens.

0 comments · 0 votesOpen discussion

Public discussion is readable by everyone. Sign in to comment, reply, or vote.

No comments yet. Start the discussion.

CURRENT ASSESSMENT · REVISION 1
TOWARD DOOM68confidence 98/100

Why it moved the index

The model combined frontier-level multimodal performance with a step change in usable context, enabling reasoning across long videos, audio, documents and codebases.

AUDIT TRAIL

Assessment history

  1. R1
    Toward 68 · confidence 98

    New February 2024 exact frontier multimodal model preview.

    12 Aug 2026
SHARE THE FINDINGS

Share this page

DoomBench social sharing card for Google previews Gemini 1.5 Pro with one-million-token context.
  1. DoomBench assesses “Google previews Gemini 1.5 Pro with one-million-token context” as evidence moving toward doom, with magnitude 68 and confidence 98 out of 100 in the capability gains category.

  2. The DoomBench assessment of “Google previews Gemini 1.5 Pro with one-million-token context” is based on reporting from Google and records the editorial rationale, source quality, attribution, and revision history.

  3. DoomBench summarizes “Google previews Gemini 1.5 Pro with one-million-token context” as follows: Google introduced Gemini 1.5 Pro in limited private preview, reporting performance comparable to Gemini 1.0 Ultra and experimental...

    https://www.doombench.com/news/google-previews-gemini-1-5-pro-with-one-million-token-context-2024-02-15