One-click GGUF quantization benchmarks & VRAM calculator on every Model Card
Display an interactive VRAM & tokens-per-second estimator for Apple Silicon, RTX 4090, and A100 directly on model repository headers across Q4_K_M, Q8, and FP16 weights.
AI & Cloud Product Roadmap · 4 Active Feature Proposals · 16,109 Priority Votes
Hugging Face is a leading AI & Cloud platform tracked on CrowdFeatures. Engineers, product managers, and power users use this board to propose new Hugging Face capabilities, upvote high-priority roadmap items, and verify features that have already launched in production.
Display an interactive VRAM & tokens-per-second estimator for Apple Silicon, RTX 4090, and A100 directly on model repository headers across Q4_K_M, Q8, and FP16 weights.
Allow side-by-side comparison of models on standardized benchmarks with custom evaluation datasets. Currently comparing models requires running separate evaluations manually.
Large organizations need model versioning, approval workflows, and audit trails. Build enterprise governance features for managing ML models across teams.
GPU availability on Spaces is unpredictable. Show real-time GPU queue depth, estimated wait times, and allow reserving GPU slots for production inference endpoints.