685B MoE variant of V3.2 with DSA, MIT license; reasoning-only tier with gold-medal IMO 2025/IOI 2025 performance and no tool-calling support.
@hn_73e440
about 1 month ago
I'm downloading DeepSeek-V3.2-Speciale now at FP8 (reportedly Gold-medal performance in the 2025 International Mathematical Olympiad and International Olympiad in Informatics). It will fit in system RAM, and as its mixture of experts and the experts are not too large, I can at least run it. Token/second speed will be slower, but as system memory bandwidth is somewhere around 5-600Gb/s, so it should feel OK.
@hn_d0a1fd
2 months ago
I stick with models I can run on VRAM, but DeepSeek Speciale have the best reasoning capabilities of the models I can actually run (https://huggingface.co/deepseek-ai/DeepSeek-V3.2-Speciale). What hardware can you access? I have Deepseek etc, but inferencing on DDR5 would take about 2-3 weeks for a simple scan. I think this works best with dense models, but it also seems ok with MoE. @everyone: Can someone hook me up with Nvidia sponsorship?