launches/olmo-hybrid
Launch
O

Olmo Hybrid

7B open model mixing transformers and linear RNNs

Sector
olmo-hybrid
O
screenshot pending

About Olmo Hybrid

Olmo Hybrid is a fully open 7B model that combines transformer attention with linear RNN layers. Utilizing a 3:1 pattern of Gated DeltaNet to attention, it matches the accuracy of Olmo 3 on MMLU while using 49% fewer tokens.

Traction

Product Hunt105 upvotes · Mar 7, 2026
Data sourced from Product Hunt