KALAVAI - Cooperative Decentralized LLM Training Specialist models from arXiv 2603.22755. Cross-lingual, private-domain, and English specialists at 410M, 1B, and Qwen-2.5-1.5B scale. mechramc/kalavai-cross-lingual-code-specialist-seed137 Text Generation • 0.4B • Updated Mar 30 • 19 mechramc/kalavai-cross-lingual-code-specialist-seed2026 Text Generation • 0.4B • Updated Apr 1 • 11 mechramc/kalavai-cross-lingual-tamil-specialist-seed137 Text Generation • 0.4B • Updated Mar 30 • 21 mechramc/kalavai-cross-lingual-tamil-specialist-seed2026 Text Generation • 0.4B • Updated Mar 30 • 7
KALAVAI - Cooperative Decentralized LLM Training Specialist models from arXiv 2603.22755. Cross-lingual, private-domain, and English specialists at 410M, 1B, and Qwen-2.5-1.5B scale. mechramc/kalavai-cross-lingual-code-specialist-seed137 Text Generation • 0.4B • Updated Mar 30 • 19 mechramc/kalavai-cross-lingual-code-specialist-seed2026 Text Generation • 0.4B • Updated Apr 1 • 11 mechramc/kalavai-cross-lingual-tamil-specialist-seed137 Text Generation • 0.4B • Updated Mar 30 • 21 mechramc/kalavai-cross-lingual-tamil-specialist-seed2026 Text Generation • 0.4B • Updated Mar 30 • 7
Running Repro - Solving Imperfect-Recall Games via Sum-of-Squares Optimization 🎯 Explore and manage experiment logs in an interactive web logbook