BeNice! Closing Looping’s Inference-Memory Gap via KV-Sharing SandyResearch/benice-1.3b Updated 18 days ago • 47 SandyResearch/benice-770m Updated 18 days ago • 36 SandyResearch/benice-370m Updated 18 days ago • 40 SandyResearch/benice-140m Updated 18 days ago • 55
Parcae A family of stable looped models SandyResearch/parcae-140m Text Generation • Updated Apr 16 • 707 • 3 SandyResearch/parcae-370m Text Generation • Updated Apr 16 • 290 • 3 SandyResearch/parcae-770m Text Generation • Updated Apr 16 • 200 • 4 SandyResearch/parcae-1.3b Text Generation • Updated Apr 16 • 454 • 8
BeNice! Closing Looping’s Inference-Memory Gap via KV-Sharing SandyResearch/benice-1.3b Updated 18 days ago • 47 SandyResearch/benice-770m Updated 18 days ago • 36 SandyResearch/benice-370m Updated 18 days ago • 40 SandyResearch/benice-140m Updated 18 days ago • 55
Parcae A family of stable looped models SandyResearch/parcae-140m Text Generation • Updated Apr 16 • 707 • 3 SandyResearch/parcae-370m Text Generation • Updated Apr 16 • 290 • 3 SandyResearch/parcae-770m Text Generation • Updated Apr 16 • 200 • 4 SandyResearch/parcae-1.3b Text Generation • Updated Apr 16 • 454 • 8