Meta 🇺🇸 · Llama 3.2

Llama 3.2 3B Instruct

The instruction-tuned three-billion-parameter Llama 3.2 checkpoint for multilingual edge assistants, tool calls, summarization, and local agentic applications.

DOOM SCORE49.6out 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 49.6

Competitive small-model instruction following and tool use create meaningful edge agency, while downloadable weights and broad hardware support maximize distribution and reduce provider control.

Capability44
Autonomy15
Deployment100
Misuse potential44
Control difficulty59
MODEL-ATTRIBUTED EVIDENCE

News tied to Llama 3.2 3B Instruct

The model score of 49.6 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+0.02

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.

Capability TOWARD

Meta releases Llama 3.2 vision and edge models

Meta released eight Llama 3.2 checkpoints spanning 1B and 3B text models for on-device tool use and 11B and 90B vision models, with downloadable weights and immediate cloud, device, and local ecosystem support.

Full item contribution
+0.19
Llama 3.2 3B Instruct equal share
+0.02
Read assessment →
AUDIT TRAIL

Model score history

  1. R3
    Doom Score 49.6

    Exact-version evidence chronology replayed after run doombench-hourly-news-20260812-142416 under temporal-monthly-pressure-v4.

    12 Aug 2026
  2. R2
    Doom Score 49.8

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

    12 Aug 2026
  3. R1
    Doom Score 50.5

    New exact open Llama 3.2 instruction checkpoint verified by Meta's dated launch.

    12 Aug 2026
SHARE THE FINDINGS

Share this page

DoomBench social sharing card for Llama 3.2 3B Instruct.
  1. Llama 3.2 3B Instruct by Meta has a DoomBench model risk score of 49.6 out of 100, based on five transparent version-specific dimensions rather than the overall index.

  2. Llama 3.2 3B Instruct's highest current DoomBench dimension is deployment at 100.0 out of 100; the profile publishes every component score and its editorial rationale.

  3. DoomBench links 1 source-backed evidence item to Llama 3.2 3B Instruct, while keeping the model's risk profile separate from each item's contribution to the live Doom Index.

    https://www.doombench.com/models/meta-llama-3-2-3b-instruct