Google DeepMind · Gemini 2.0

Gemini 2.0 Flash Experimental

READER SUMMARY

An experimental multimodal model with native image and audio output, search and code tools, function calling, planning and low-latency API and Gemini-app access.

DOOM SCORE71.4out of 100model risk profile, not the overall index

CURRENT ASSESSMENT · REVISION 1

Why this model scores 71.4

The model exceeds Gemini 1.5 Pro on key benchmarks and materially raises native tool use and multimodal agency. API and global assistant access create high deployment, while experimental gating, hosted controls and human confirmation on sensitive agent actions moderate control difficulty.

Capability77

Autonomy52

Deployment88

Misuse potential64

Control difficulty55

0 comments · 0 votesOpen discussion

No comments yet. Start the discussion.

MODEL-ATTRIBUTED EVIDENCE

News tied to Gemini 2.0 Flash Experimental

The model score of 71.4 rates this model's risk profile. The overall Doom Index of 63.2 measures the recalibrated complete evidence corpus. These values answer different questions.

NET MODEL-ATTRIBUTED INDEX CONTRIBUTION+0.05

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.

11 Dec 2024AutonomyTOWARD

Google releases Gemini 2.0 Flash Experimental

Google released Gemini 2.0 Flash Experimental to developers and Gemini users with native tool use, multimodal output, planning and support for agent prototypes including Mariner and Jules.

Full item contribution: +0.05
Gemini 2.0 Flash Experimental equal share: +0.05

Read assessment →

AUDIT TRAIL

Model score history

R1
Doom Score 71.4
Initial exact source-backed profile for Gemini 2.0 Flash Experimental.
12 Aug 2026