DeepSeek · DeepSeek-V4

DeepSeek-V4-Flash

READER SUMMARY

Smaller open-weight DeepSeek-V4 tier offering one-million-token context, inexpensive API access, and simple-agent performance near the Pro tier.

DOOM SCORE78.1out of 100model risk profile, not the overall index
CURRENT ASSESSMENT · REVISION 1

Why this model scores 78.1

Lower breadth than V4-Pro is offset by very high accessibility, low cost, open weights, and integration into common agent frameworks.

Capability74
Autonomy70
Deployment99
Misuse potential68
Control difficulty84
MODEL-ATTRIBUTED EVIDENCE

News tied to DeepSeek-V4-Flash

The model score of 78.1 rates this model's risk profile. The overall Doom Index of 62.1 measures the recalibrated complete evidence corpus. These values answer different questions.

NET MODEL-ATTRIBUTED INDEX CONTRIBUTION+0.06

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.

AUDIT TRAIL

Model score history

  1. R1
    Doom Score 78.1

    New exact open-weight tier with materially distinct size, cost, performance, and deployment profile.

    11 Aug 2026