Kiyo Kiyo: State-of-the-art models trained from scratch on vast amounts of web tokens, educational text, and code. DedeProGames/Kiyo-230M-Instruct Text Generation • 0.2B • Updated 7 days ago • 604 • 7 DedeProGames/Kiyo-230M-Preview Text Generation • 0.2B • Updated 2 days ago • 531 • 6 DedeProGames/Kiyo-65M Text Generation • 65M • Updated 2 days ago • 805 • 3
GPT-U GPT-U: small Llama-style base models trained from scratch on web, educational text and code, with fully open scripts, logs and evals. DedeProGames/GPT-U-20M-Instruct Text Generation • 20.5M • Updated 6 days ago • 499 • 5 DedeProGames/GPT-U-20M Text Generation • 20.5M • Updated 26 days ago • 1.09k • 2
Kiyo Kiyo: State-of-the-art models trained from scratch on vast amounts of web tokens, educational text, and code. DedeProGames/Kiyo-230M-Instruct Text Generation • 0.2B • Updated 7 days ago • 604 • 7 DedeProGames/Kiyo-230M-Preview Text Generation • 0.2B • Updated 2 days ago • 531 • 6 DedeProGames/Kiyo-65M Text Generation • 65M • Updated 2 days ago • 805 • 3
GPT-U GPT-U: small Llama-style base models trained from scratch on web, educational text and code, with fully open scripts, logs and evals. DedeProGames/GPT-U-20M-Instruct Text Generation • 20.5M • Updated 6 days ago • 499 • 5 DedeProGames/GPT-U-20M Text Generation • 20.5M • Updated 26 days ago • 1.09k • 2
pinned Running Agents 17 SLM Tetris Arena 🧱 Tiny decoder-only LMs play Tetris zero-shot, ranked by Elo
DedeProGames/SoralASR-23M-Preview Automatic Speech Recognition • 23.1M • Updated 29 days ago • 120 • 1