view post Post 190 Hey if you're reading this and happen to be one of the guys training frontier llms, please penalize 404 urls in your reward functions. Happens too often that these models memorize / make up non-existing url paths and get away with it See translation 1 reply · 🔥 1 1 + Reply
Tongyi-Zhiwen/QwenLong-L1.5-30B-A3B Text Generation • 31B • Updated about 18 hours ago • 14.9k • 162
nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 Text Generation • 32B • Updated 1 day ago • 299k • 551
RedHatAI/Meta-Llama-3.1-8B-Instruct-FP8-dynamic Text Generation • 8B • Updated 29 days ago • 37k • 9
NVIDIA Nemotron v3 Collection Open, Production-ready Enterprise Models • 6 items • Updated 10 days ago • 116
TeichAI/gemini-3-pro-preview-high-reasoning-1000x Viewer • Updated about 1 month ago • 1.02k • 1.48k • 56
BarraHome/vmware-embeddings-large-v1 Sentence Similarity • 0.1B • Updated about 1 month ago • 23
BarraHome/vmware-embeddings-large-v1 Sentence Similarity • 0.1B • Updated about 1 month ago • 23
RedHatAI/Llama-3.1-8B-Instruct-speculator.eagle3 Text Generation • 1.0B • Updated 21 days ago • 6.41k • 1
Running 1 Quantization Formats And Cuda Compute Capability Support 🧠1 Quantization Formats & CUDA Compute Capability Support
Running 1 Quantization Formats And Cuda Compute Capability Support 🧠1 Quantization Formats & CUDA Compute Capability Support
Running 1 Quantization Formats And Cuda Compute Capability Support 🧠1 Quantization Formats & CUDA Compute Capability Support