A post announcing the update says DeepSeek-V3.2 inspection time fell from 9.2 seconds to 0.45 seconds.
A post announcing the update says inspecting model tensors on the Hugging Face Hub is up to 20× faster. It lists inspection times falling from 9.2 to 0.45 seconds for DeepSeek-V3.2, 4.0 to 0.57 seconds for GLM-5.3, and 1.2 to 0.27 seconds for Qwen2.5-72B.
6K
2 Sources, first seen ago
@victormustarWe made inspecting models tensors on the Hugging Face Hub up to 20× faster ⚡ DeepSeek-V3.2: 9.2s → 0.45s GLM-5.3: 4.0s → 0.57s Qwen2.5-72B: 1.2s → 0.27s
@ClementDelangueRT @victormustar: We made inspecting models tensors on the Hugging Face Hub up to 20× faster ⚡ DeepSeek-V3.2: 9.2s → 0.45s GLM-5.3: 4.0s →…@victormustarWe made inspecting models tensors on the Hugging Face Hub up to 20× faster ⚡ DeepSeek-V3.2: 9.2s → 0.45s GLM-5.3: 4.0s → 0.57s Qwen2.5-72B: 1.2s → 0.27s@ClementDelangueRT @victormustar: We made inspecting models tensors on the Hugging Face Hub up to 20× faster ⚡ DeepSeek-V3.2: 9.2s → 0.45s GLM-5.3: 4.0s →…—
Not ranked yet
—
Not ranked yet
