Tighten 2.1.1 docs

CHANGELOG.md

@@ -16,7 +16,7 @@ Updated the Video Refiner rating categories so feedback can separate missing det
 
 Reduced repeated category feedback prompts after the user has already answered enough about the same concept.
 
-Refreshed README and refiner docs for the promoted `main` release.
+Refreshed README and refiner docs for 2.1.1.
 
 ### Fixed
 
@@ -36,7 +36,7 @@ The new workflow uses base LoRA weights on the first run for a prompt, then lets
 
 ### Changed
 
-Promoted the `FunPack Video Refiner` development workflow to `main` and renamed the visible refiner title from `FunPack Gemma Embedding Refiner` to `FunPack Video Refiner` while keeping the old node key as a compatibility alias.
+Renamed the visible refiner title from `FunPack Gemma Embedding Refiner` to `FunPack Video Refiner`. The old node key is still available as a compatibility alias.
 
 Split the old single `funpack.py` implementation into focused modules:
 

README.md

@@ -2,13 +2,13 @@
 
 A set of ComfyUI nodes for experimenting with video generation workflows based on WAN, HunyuanVideo, LTX, and similar models.
 
-## What's New in 2.1.1
+## Updates in 2.1.1
 
-Version 2.1.1 brings the current Video Refiner work from `dev` to `main`.
+New rating logic for `FunPack Video Refiner`, plus a new CLIP Vision Combine node.
 
-`FunPack Gemma Embedding Refiner` now appears in ComfyUI as `FunPack Video Refiner`. Old workflows still load, because the original node key is kept as an alias.
+`FunPack Gemma Embedding Refiner` now appears in ComfyUI as `FunPack Video Refiner`. Old workflows still load because the original node key is kept as an alias.
 
-The main changes are:
+Highlights:
 
 - clearer rating choices, including `-Just forget it-` for bad runs you do not want saved
 - concept feedback questions that can ask what kind of thing a prompt phrase is: concept, style, quality, character, details, or general text
@@ -79,6 +79,7 @@ Per-node documentation is available in the [`docs`](docs) folder.
 Start with:
 
 - [`docs/FunPackGemmaEmbeddingRefiner.md`](docs/FunPackGemmaEmbeddingRefiner.md) for `FunPack Video Refiner`
+- [`docs/FunPackClipVisionOutputCombine.md`](docs/FunPackClipVisionOutputCombine.md) for CLIP Vision output combining
 - [`docs/FunPackSaveRefinementLatent.md`](docs/FunPackSaveRefinementLatent.md) for latent references
 - [`docs/FunPackLoraWorkflow.md`](docs/FunPackLoraWorkflow.md) for the LoRA/refiner helper workflow
 

docs/FunPackClipVisionOutputCombine.md

@@ -0,0 +1,24 @@
+# FunPack CLIP Vision Output Combine
+
+This node combines up to four `CLIP_VISION_OUTPUT` inputs into one output.
+
+Use it when a workflow has several CLIP Vision encodes and the next node only accepts one CLIP Vision output.
+
+## Parameters
+
+**clip_vision_output1**: Required CLIP Vision output.
+
+**clip_vision_output2, clip_vision_output3, clip_vision_output4**: Optional extra CLIP Vision outputs.
+
+**method**: How tensor fields are combined:
+
+- `mean`: average value
+- `median`: median value
+- `maximum`: highest value
+- `minimum`: lowest value
+
+All tensor fields being combined must have the same shape. Non-tensor fields are copied from the first input.
+
+## Output
+
+**clip_vision_output**: Combined CLIP Vision output.