Document FC2 sidecar and modifications

Document artifact hashes, exact-output performance measurements, automatic source validation, attribution, and the MiniMax-required modification notice.

NOTICE

@@ -0,0 +1,10 @@
+MiniMax H3 is licensed under the MiniMax H3 Community License Agreement, Copyright © 2026 MiniMax. All Rights Reserved.
+
+Modification notice
+
+PulpCut transferred the credited FL2VA Turbo adapter to the Ref2VA transformer
+and requantized it to the documented INT8 ConvRot format. PulpCut also derived
+the optional FC2 input-major sidecar by transposing the storage of the 50 INT8
+mlp.fc2.weight tensors. The sidecar does not change tensor values, scales,
+reference conditioning, or model behavior. See README.md for sources, hashes,
+measurements, and reproduction instructions.

README.md

@@ -16,25 +16,45 @@ base_model: MiniMaxAI/MiniMax-H3
 pretty_name: PulpCut MiniMax H3 Ref2VA Turbo INT8 ConvRot
 ---
 
-# MiniMax H3 Ref2VA Turbo · pruned INT8 ConvRot (single file)
+# MiniMax H3 Ref2VA Turbo · pruned INT8 ConvRot
 
 ## What this repository is
 
-A single-file MiniMax H3 **Ref2VA** diffusion transformer — the omni-reference
-variant, which conditions generation on ordered reference images — with the
-lightx2v **turbo step-distillation merged into the weights**, quantized in the
-same pruned **INT8 ConvRot** layout as the Comfy-Org release. It is a drop-in
-replacement for `minimax_h3_ref2va_pruned_int8_convrot.safetensors` in any
-runtime that reads the optimized INT8 single-file layout — including
+An optimized MiniMax H3 **Ref2VA** package centered on the omni-reference
+diffusion transformer, which conditions generation on ordered reference
+images. It has the lightx2v **turbo step-distillation merged into the weights**
+and uses the same pruned **INT8 ConvRot** layout as the Comfy-Org release. The
+primary transformer is a drop-in replacement for
+`minimax_h3_ref2va_pruned_int8_convrot.safetensors` in any runtime that reads
+the optimized INT8 layout — including
 [H3ddle](https://github.com/AlexanderIstomin/h3ddle), the open-source native
 macOS app it was built for.
 
-This file is **not a standalone model**. It needs the rest of the optimized
+The transformer is **not a standalone model**. It needs the rest of the optimized
 package (Qwen3-VL-32B INT8 text encoder, video/audio VAEs, tokenizer) from
 [Comfy-Org/MiniMax-H3](https://huggingface.co/Comfy-Org/MiniMax-H3), and the
 FL2VA transformer alongside it if you also want prompt-only and keyframe
 generation.
 
+## H3ddle FC2 performance sidecar
+
+`minimax_h3_ref2va_pruned_turbo_int8_convrot_fc2_input_major.safetensors`
+is an optional H3ddle performance sidecar derived from the transformer in
+this repository. It contains only the 50 INT8 `mlp.fc2.weight` matrices, with
+their storage transposed from `[output, input]` to `[input, output]`. Values,
+quantization scales, reference conditioning, and model behavior are unchanged.
+
+H3ddle installs it beside the transformer and selects it automatically after
+validating its format version, source file size, exact source-header
+fingerprint, and all 50 tensor schemas. It cannot silently be used with a
+different checkpoint. The original transformer remains available as the
+fallback and for runtimes that do not understand the sidecar.
+
+A real 512-class, 50-block Ref2VA parity run produced identical baseline and
+sidecar hashes: video `85a5ccfc5a4d8075`, audio `731e24ae9dc2e7ec`.
+The matched cold pair measured 43.753 seconds without the sidecar and 40.736
+seconds with it; the broader FL2VA A/B/B/A benchmark measured a 7.15% gain.
+
 ## Why this merge was made and republished
 
 Every published turbo LoRA for MiniMax H3 targets the **FL2VA** transformer.
@@ -100,7 +120,7 @@ Derivative of MiniMax H3 weights; the
 [MiniMax H3 Community License Agreement](https://huggingface.co/MiniMaxAI/MiniMax-H3/blob/939557dc319dd91227e30195a763f272ba7f8765/LICENSE)
 applies. By downloading you agree to its terms.
 
-## What this file is used for in H3ddle
+## What these files are used for in H3ddle
 
 H3ddle installs it as a managed model alongside the reference-capable package:
 the app verifies the SHA-256 below, reuses the shared package files it already
@@ -119,6 +139,7 @@ of the MiniMax H3 Community License apply unchanged.
 | File | Bytes | SHA-256 |
 |---|---|---|
 | `minimax_h3_ref2va_pruned_turbo_int8_convrot.safetensors` | 20,970,379,854 | `e64cef63bc2785bcd72e6103c52aa78c6cd2c4f9870a7ce79675083fd65cf2e7` |
+| `minimax_h3_ref2va_pruned_turbo_int8_convrot_fc2_input_major.safetensors` | 3,853,522,260 | `0ad6a5673abdf842c39d4d8de7c34c971a420b64bd5f79eb6f4331c5bfb5cd97` |
 
 ## Reproducibility references
 
@@ -127,6 +148,9 @@ The conversion is a single dependency-free Python script,
 in the H3ddle repository, including the strength-0 self-check used to validate
 the pipeline against the official file.
 
+The FC2 sidecar is reproducible with
+[`Scripts/optimize-h3-fc2-sidecar.py`](https://github.com/AlexanderIstomin/h3ddle/blob/main/Scripts/optimize-h3-fc2-sidecar.py).
+
 ## Contact
 
 Open an issue in the [H3ddle repository](https://github.com/AlexanderIstomin/h3ddle/issues).