Reaction
A plan to make gpt oss 120b match luna's performance
A user lists new RL environments from mimo and slime as parts of the proposed path.
TLDR
A user sketches a plan to take new RL environments from mimo, use gpt oss 120b and slime, and aim to make gpt oss 120b match luna's performance.
Combined views
1.8K
1 Source, first seen 4h ago
35 likes