Grok 3
READER SUMMARYSpaceXAI's flagship third-generation beta model, combining extensive pretraining with reinforcement-learned test-time reasoning and improved mathematics, coding and instruction following.
Why this model scores 74.4
Large capability gains and minutes-long self-correcting reasoning raise capability and autonomy. Early user rollout makes deployment material but below generally available services. Broad X-linked access and strong coding increase misuse exposure, while hosted access limits downstream control difficulty.
0 comments · 0 votes
Sign in to join the discussion →
No comments yet. Start the discussion.
News tied to Grok 3
The model score of 74.4 rates this model's risk profile. The overall Doom Index of 63.2 measures the recalibrated complete evidence corpus. These values answer different questions.
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.
SpaceXAI unveils Grok 3 and Grok 3 mini reasoning models
SpaceXAI unveiled Grok 3 and Grok 3 mini, beta reasoning models trained with large-scale reinforcement learning and ten times the compute of its prior generation, with user rollout beginning.
- Full item contribution
- +0.04
- Grok 3 equal share
- +0.02
Model score history
- R1Doom Score 74.4
Initial exact profile for the February 2025 flagship Grok 3 beta tier.
12 Aug 2026