helloericsf | Dark Hacker News

1.	Context Engineering for AI Agents: Lessons(manus.im) 120 points by helloericsf 236 days ago \| 4 comments
2.	Context Engineering for AI Agents: Lessons(manus.im) 3 points by helloericsf 303 days ago \| 0 comments
3.	Better than DeepSeek R1? MiniMax-M1:open-weight hybrid-attention reasoning model(huggingface.co) 6 points by helloericsf 335 days ago \| 0 comments
4.	kit - Code Intelligence Toolkit(github.com) 1 points by helloericsf 1 year ago \| 0 comments
5.	DeepSeek Open Source Optimized Parallelism Strategies, 3 repos(github.com) 103 points by helloericsf 1 year ago \| 8 comments
6.	DeepSeek Open Source DeepGEMM – FP8 GEMM Library(300 lines for 1350+ FP8 TFLOPS)(twitter.com) 4 points by helloericsf 1 year ago \| 1 comment
7.	Alibaba Open Source Large-Scale Video Generative Models: Wan2.1(twitter.com) 8 points by helloericsf 1 year ago \| 2 comments
8.	DeepSeek open source DeepEP – library for MoE training and Inference(github.com) 536 points by helloericsf 1 year ago \| 71 comments
9.	DeepSeek Open Source FlashMLA – MLA Decoding Kernel for Hopper GPUs(github.com) 441 points by helloericsf 1 year ago \| 108 comments
10.	New Qwen2.5-Max Outperforms DeepSeek V3 in Benchmarks(twitter.com) 3 points by helloericsf 1 year ago \| 2 comments
11.	Longest context up to 4M, MiniMax-01 hybrid 456B Open source model(github.com) 19 points by helloericsf 1 year ago \| 1 comment