deepseek-ai/DeepSeek-V4-Pro Text Generation • 862B • Updated about 1 hour ago • 138k • • 2.94k
mlx-community/Qwen3-TTS-12Hz-1.7B-CustomVoice-8bit Text-to-Speech • 0.8B • Updated Jan 26 • 5.21k • 22
Kimi Linear: An Expressive, Efficient Attention Architecture Paper • 2510.26692 • Published Oct 30, 2025 • 132
Running on CPU Upgrade Featured 3.13k The Smol Training Playbook 📚 3.13k The secrets to building world-class LLMs