Google DeepMind · Gemma

Gemma 2B IT

READER SUMMARY

Downloadable two-billion-parameter instruction-tuned Gemma checkpoint optimized for conversational and directed tasks.

DOOM SCORE43.1out of 100model risk profile, not the overall index

CURRENT ASSESSMENT · REVISION 1

Why this model scores 43.1

Instruction tuning made the compact open-weight model easier to direct for useful and harmful tasks, while small size enabled exceptionally broad local deployment.

Capability34

Autonomy5

Deployment96

Misuse potential37

Control difficulty51

0 comments · 0 votesOpen discussion

No comments yet. Start the discussion.

MODEL-ATTRIBUTED EVIDENCE

News tied to Gemma 2B IT

The model score of 43.1 rates this model's risk profile. The overall Doom Index of 61.2 measures the recalibrated complete evidence corpus. These values answer different questions.

NET MODEL-ATTRIBUTED INDEX CONTRIBUTION+0.01

Each article's leave-one-out 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 are recalculated from their technical profile and related news, but never feed back into the overall index.

21 Feb 2024RaceTOWARD

Google releases four open-weight Gemma 2B and 7B variants

Google released pretrained and instruction-tuned Gemma models at 2B and 7B sizes with downloadable weights, broad framework support and deployment options spanning laptops, workstations and cloud infrastructure.

Full item contribution: +0.03
Gemma 2B IT equal share: +0.01

Read assessment →

AUDIT TRAIL

Model score history

R1
Doom Score 43.1
New exact Gemma 2B instruction-tuned profile.
12 Aug 2026