DeepReinforce releases Ornith-1.0 open-source coding models
DeepReinforce released Ornith-1.0, open-source coding models from 9B to 397B that learn both scaffolds and solutions during RL. With layered anti-reward-hacking safeguards, the family posts strong coding benchmark results across model sizes.
🗞 #sponsored @testingcatalog
Post #8466
1.27K