OctoThinker

community

https://github.com/GAIR-NLP/OctoThinker

GAIR-NLP

AI & ML interests

None defined yet.

Recent Activity

koalazf99 authored a paper 5 days ago

Revisiting Reinforcement Learning for LLM Reasoning from A Cross-Domain Perspective

koalazf99 published a model 2 months ago

OctoThinker/OctoThinker-8B-Hybrid-Base

koalazf99 published a model 2 months ago

OctoThinker/OctoThinker-8B-Short-Base

View all activity

Organization Card

Community About org cards

🐙 OctoThinker is led by GAIR

🎯 Our Goal: To reshape the pre-training trajectory so models scale better under RL.

models 16

OctoThinker/OctoThinker-8B-Hybrid-Base

Updated Apr 24 • 12 • 2

OctoThinker/OctoThinker-8B-Short-Base

Updated Apr 24 • 46

OctoThinker/OctoThinker-8B-Long-Base

Updated Apr 24 • 11

OctoThinker/OctoThinker-3B-Short-Zero

Updated Apr 23 • 8

OctoThinker/OctoThinker-3B-Hybrid-Zero

Updated Apr 23 • 32

OctoThinker/OctoThinker-1B-Long-Zero

Updated Apr 23 • 7

OctoThinker/OctoThinker-1B-Hybrid-Zero

Updated Apr 23 • 7

OctoThinker/OctoThinker-1B-Short-Zero

Updated Apr 23 • 41

OctoThinker/Llama3.2-3B-Zero

Updated Apr 22 • 11

OctoThinker/OctoThinker-3B-Long-Zero

Updated Apr 22 • 8

datasets 0

None public yet