9 results

sm_86

inall models datasets apps agents people projects

Models

All models matching “sm_86”

gbueno86 /Meta-LLama-3-Cat-Smaug-LLama-70btext-generationtransformersen1 likes8.6k downloads2y agoHugging Face wei231 /ternary-bonsai-3080-sm86-ninferninferzh0 likes1.3k downloads2d agoHugging Face featherless-ai-quants /gbueno86-Meta-LLama-3-Cat-Smaug-LLama-70b-GGUFtext-generation0 likes457 downloads1y agoHugging Face nith3n /qwen3-4b-instruct-2507-qairt-sm8650-w4a160 likes373 downloads1mo agoHugging Face zLLM-Lab /zllm-minicpm5-2b-qnn-sm86350 likes113 downloads14d agoHugging Face leslie721007 /babylm-strict-small-coherent86fill-masktransformersen1 likes33 downloads22d agoHugging Face leslie721007 /babylm-strict-small-coherent86-alpha075fill-masktransformersen1 likes29 downloads21d agoHugging Face satish860 /sms_detection_algorithmtext-classificationtransformers0 likes20 downloads4y agoHugging Face

Datasets

All datasets matching “sm_86”

Relativ3pa1n /dsv4-flash-sm86-8x3090 DeepSeek-V4-Flash on 8x RTX 3090 (SM86): 262K context, 120 tok/s aggregate Serving recipes, launch wrappers, and the measured throughput/context ladder for running a W4A16 DeepSeek-V4-Flash-class model on 8x RTX 3090 (SM 8.6, 24 GiB each) with CUDA graph decode, FlashInfer sparse MLA, Marlin MoE, and compressed hybrid KV. The ladder Concurrent sequences amortize the TP8 allreduce that dominates each decode step, so aggregate throughput scales near-linear while… See the full description on the dataset page: https://huggingface.co/datasets/Relativ3pa1n/dsv4-flash-sm86-8x3090.0 likes90 downloads28d agoHugging Face