Meta · Llama 3.1

Llama 3.1 405B Instruct

READER SUMMARY

Meta's frontier-scale 405B instruction model with 128K context, multilingual reasoning, tool use, downloadable weights and mass partner distribution.

DOOM SCORE68.6out of 100model risk profile, not the overall index

CURRENT ASSESSMENT · REVISION 1

Why this model scores 68.6

Near-frontier general capability, tool use and direct instruction following combine with downloadable weights and broad hosting, creating high misuse potential and persistent control difficulty despite compute requirements.

Capability80

Autonomy24

Deployment82

Misuse potential75

Control difficulty77

0 comments · 0 votesOpen discussion

No comments yet. Start the discussion.

MODEL-ATTRIBUTED EVIDENCE

News tied to Llama 3.1 405B Instruct

The model score of 68.6 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+0.01

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.

23 Jul 2024CapabilityTOWARD

Meta releases frontier-scale Llama 3.1 open models

Meta released Llama 3.1 in 8B, 70B and 405B base and instruction tiers with 128K context, multilingual support, tool use, downloadable weights and day-one distribution through more than 25 partners.

Full item contribution: +0.05
Llama 3.1 405B Instruct equal share: +0.01

Read assessment →

AUDIT TRAIL

Model score history

R1
Doom Score 68.6
New exact Llama 3.1 405B instruction checkpoint.
12 Aug 2026