view post Post 2097 Mistral's new SOTA coding models Devstral 2 can now be Run locally! (25GB RAM) 🐱We fixed the chat template, so performance should be much better now!24B: unsloth/Devstral-Small-2-24B-Instruct-2512-GGUF123B: unsloth/Devstral-2-123B-Instruct-2512-GGUF🧡Step-by-step Guide: https://docs.unsloth.ai/models/devstral-2 See translation 🔥 8 8 🚀 5 5 ❤️ 3 3 🤗 2 2 + Reply
A Survey on Hypothesis Generation for Scientific Discovery in the Era of Large Language Models Paper • 2504.05496 • Published Apr 7, 2025
KaVa: Latent Reasoning via Compressed KV-Cache Distillation Paper • 2510.02312 • Published Oct 2, 2025 • 1
Beyond Recognition: Evaluating Visual Perspective Taking in Vision Language Models Paper • 2505.03821 • Published May 3, 2025 • 24
Joint MoE Scaling Laws: Mixture of Experts Can Be Memory Efficient Paper • 2502.05172 • Published Feb 7, 2025 • 2