INTELLECT-3: A 100B+ MoE trained with large-scale RL
AI & ML interests
Commoditizing intelligence through a decentralized AI lab to advance scientific progress and human flourishing.
Recent Activity
View all activity
INTELLECT-2 is a 32 billion parameter language model with globally distributed reinforcement learning.
A collection of tasks & verifiers for reasoning datasets
-
PrimeIntellect/stackexchange-question-answering
Viewer • Updated • 313k • 178 • 11 -
PrimeIntellect/real-world-swe-problems
Viewer • Updated • 69.8k • 33 • 13 -
PrimeIntellect/synthetic-code-understanding
Viewer • Updated • 60.6k • 50 • 20 -
PrimeIntellect/verifiable-math-problems
Viewer • Updated • 777k • 85 • 21
INTELLECT-3: A 100B+ MoE trained with large-scale RL
INTELLECT-2 is a 32 billion parameter language model with globally distributed reinforcement learning.
A collection of tasks & verifiers for reasoning datasets
-
PrimeIntellect/stackexchange-question-answering
Viewer • Updated • 313k • 178 • 11 -
PrimeIntellect/real-world-swe-problems
Viewer • Updated • 69.8k • 33 • 13 -
PrimeIntellect/synthetic-code-understanding
Viewer • Updated • 60.6k • 50 • 20 -
PrimeIntellect/verifiable-math-problems
Viewer • Updated • 777k • 85 • 21
INTELLECT-1 Training dataset