The vllm-sr/d3-nano galaxy

16 models are built on vllm-sr/d3-nano: 10 quantized, 6 fine-tuned, 0 adapters and 0 merges. Together they were downloaded 21,375 times in the last 30 days.

Most downloaded direct derivatives

• ggml-org/vllm-d3-nano-GGUF (quantized), 12.8K downloads in 30 days
• Jundot/d3-nano-oQ8e (quantized), 221 downloads in 30 days
• thoddnn/d3-nano-q4 (finetune), 117 downloads in 30 days
• FastFlowLM/D3-Nano-NPU2 (finetune), 105 downloads in 30 days
• FluidInference/d3-nano-coreml (quantized), 103 downloads in 30 days
• wegotomars/d3-nano-CoreAI-ANE (finetune), 90 downloads in 30 days
• ahmedandaloes/d3-nano-mlx (quantized), 78 downloads in 30 days
• qybaihe/mude-1.0 (finetune), 54 downloads in 30 days
• Vishva007/d3-nano-W4A16-AutoRound (quantized), 50 downloads in 30 days
• Jakevin/d3-nano-ternary-mlx (quantized), 48 downloads in 30 days
• ArthurZ/d3-nano (finetune), 47 downloads in 30 days
• tensor-ai-inc/d3-nano-NVFP4 (quantized), 37 downloads in 30 days
• ahmedandaloes/d3-nano-CoreAI (quantized), 34 downloads in 30 days
• Jakevin/d3-nano-ternary-3bit-mlx (quantized), 33 downloads in 30 days
• maxffarrell/d3-nano-CoreAI (finetune), 29 downloads in 30 days

Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.