helloericsf

Posts

• Better than DeepSeek R1? MiniMax-M1:open-weight hybrid-attention reasoning model
  by helloericsfon 6/16/2025, 5:28 PMwith 0 comments
• kit - Code Intelligence Toolkit
  by helloericsfon 5/8/2025, 11:16 PMwith 0 comments
• DeepSeek Open Source Optimized Parallelism Strategies, 3 repos
  by helloericsfon 2/27/2025, 2:01 AMwith 8 comments
• DeepSeek Open Source DeepGEMM – FP8 GEMM Library(300 lines for 1350+ FP8 TFLOPS)
  by helloericsfon 2/26/2025, 1:08 AMwith 1 comments
• Alibaba Open Source Large-Scale Video Generative Models: Wan2.1
  by helloericsfon 2/25/2025, 3:03 PMwith 2 comments
• DeepSeek open source DeepEP – library for MoE training and Inference
  by helloericsfon 2/25/2025, 2:27 AMwith 71 comments
• DeepSeek Open Source FlashMLA – MLA Decoding Kernel for Hopper GPUs
  by helloericsfon 2/24/2025, 1:37 AMwith 108 comments
• New Qwen2.5-Max Outperforms DeepSeek V3 in Benchmarks
  by helloericsfon 1/28/2025, 4:08 PMwith 2 comments
• Longest context up to 4M, MiniMax-01 hybrid 456B Open source model
  by helloericsfon 1/14/2025, 7:32 PMwith 1 comments
• DeepSeek v3 beats Claude sonnet 3.5 and way cheaper
  by helloericsfon 12/26/2024, 11:47 AMwith 9 comments
• NeurIPS and Dr. Picard released statement for singling out Chinese scholars
  by helloericsfon 12/16/2024, 6:16 PMwith 2 comments
• Tencent Hunyuan-Large
  by helloericsfon 11/5/2024, 6:52 PMwith 103 comments