Upload README.md with huggingface_hub

README.md

@@ -0,0 +1,16 @@
+# FastH3 Pruned 8-Step (full NVFP4, checkpoint 300)
+
+**Full NVFP4 (W4A4), calibrated.** Every block's MLP (`ff.fc_in`, `ff.fc_out`), attention projections
+(`attn.to_q/k/v`, `attn.to_out`) and VSA compression gate (`attn.to_gate_compress`) are NVFP4, each with a **static
+activation scale max-calibrated on 1,000 prompts x all DMD steps**, stored in `transformer/nvfp4_weights.safetensors`
+as `_nvfp4_input_global_sf`. AdaLN, token refiner, embeddings and heads stay in their source precision.
+Requires an NVIDIA Blackwell GPU (sm100/sm120). This is the layout of the RTX 5090 speed configuration
+(the whole DiT stays resident on a 32 GB card).
+
+- 42-block pruned FastH3, rank-16 AdaLN, 8-step DMD (999..125), VSA 0.8, training checkpoint 300.
+- `text_encoder`: Qwen3-VL conditioner, NVFP4, 50 of 64 language layers (lossless: H3 reads hidden state 50).
+- `vae`: light VAE: 26-block LynnReal decoder + Kijai int8-convrot overlay.
+- `fastvideo_inference.json`: the sampling contract; FastVideo reads it automatically.
+- Load with `engine.quantization = {"transformer_quant": "NVFP4", "layer_profile": "h3_dit_vsa"}`.
+
+**Status: internal, private evaluation build.** License: MiniMax H3 Community License (inherited).