Google 🇺🇸 · Gemma 4

Gemma 4 31B Dense

The strongest dense Gemma 4 tier, locally deployable and adaptable for advanced reasoning, coding, and agents.

DOOM SCORE75.0out of 100model risk profile, not the overall index
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 3

Why this model scores 75.0

The strongest Gemma 4 tier combines local adaptability and agent support with higher hardware requirements that modestly constrain diffusion.

Capability83
Autonomy74
Deployment91
Misuse potential64
Control difficulty60
MODEL-ATTRIBUTED EVIDENCE

News tied to Gemma 4 31B Dense

The model score of 75.0 rates this model's risk profile. The overall Doom Index of 67.9 measures the complete temporally weighted evidence record. These values answer different questions.

NET MODEL-ATTRIBUTED INDEX CONTRIBUTION+0.05

Each article's current Doom Index contribution is divided equally among the exact models named on that article. This prevents multi-model evidence from being claimed in full on several model pages. Model risk scores use a bounded temporal offset around their technical profile, but never feed back into the overall index.

AUDIT TRAIL

Model score history

  1. R3
    Doom Score 75.0

    Exact-version evidence chronology replayed after run doombench-hourly-news-20260812-132112 under fixed-sensitivity-v3.

    12 Aug 2026
  2. R2
    Doom Score 75.2

    Full-corpus evidence recalculated after run intensive-backfill-20260811-191225 under bounded-corpus-v2.

    11 Aug 2026
  3. R1
    Doom Score 75.0

    New exact tier verified from dated primary launch.

    11 Aug 2026
SHARE THE FINDINGS

Share this page

DoomBench social sharing card for Gemma 4 31B Dense.
  1. Gemma 4 31B Dense by Google has a DoomBench model risk score of 75.0 out of 100, based on five transparent version-specific dimensions rather than the overall index.

  2. Gemma 4 31B Dense's highest current DoomBench dimension is deployment at 91.0 out of 100; the profile publishes every component score and its editorial rationale.

  3. DoomBench links 1 source-backed evidence item to Gemma 4 31B Dense, while keeping the model's risk profile separate from each item's contribution to the live Doom Index.

    https://www.doombench.com/models/google-deepmind-gemma-4-31b-dense