Follow the rename, and say which 8 GB the name means

README.md

@@ -18,9 +18,13 @@ tags:
 base_model: Lightricks/LTX-2.5
 ---
 
-[![Haverbex LTX2.5-4bit](https://huggingface.co/topabaem/LTX-2.5-Text-Encoder-4bit/resolve/main/banner.png)](https://buymeacoffee.com/choijjs83q)
+[![Haverbex LTX2.5-4bit](https://huggingface.co/topabaem/LTX-2.5-Text-Encoder-4bit-8GB/resolve/main/banner.png)](https://buymeacoffee.com/choijjs83q)
 
-# LTX-2.5 Text Encoder — 4-bit
+# LTX-2.5 Text Encoder — 4-bit, 8 GB
+
+> **The 8 GB in the name is the file**, not the card. On disk this is 8.46 GB
+> against the original's 26.264 GB. Running it needs about **9.7 GB of VRAM** —
+> an 8 GB card is not enough. See [Memory](#memory--what-the-card-actually-needs).
 
 > **EN** — I'm a student researching ML quantization. Getting this one done
 > burned through so much in server bills that from here on I'll only be able to
@@ -115,25 +119,25 @@ All five deterministic pairs at a glance — BF16 left, 4-bit right, one row per
 prompt. Rows 1, 3, 4 and 5 hold together; row 2, the dune, is where the
 compression is visible.
 
-![BF16 against 4-bit, five deterministic pairs](https://huggingface.co/topabaem/LTX-2.5-Text-Encoder-4bit/resolve/main/samples/comparison-sheet.png)
+![BF16 against 4-bit, five deterministic pairs](https://huggingface.co/topabaem/LTX-2.5-Text-Encoder-4bit-8GB/resolve/main/samples/comparison-sheet.png)
 
 ### Watch the two that matter
 
 **Forge — the pair holding together.** Deterministic sampler, same seed. Left is
 BF16, right is 4-bit.
 
-<video controls width="100%" src="https://huggingface.co/topabaem/LTX-2.5-Text-Encoder-4bit/resolve/main/samples/compare-det-02.mp4"></video>
+<video controls width="100%" src="https://huggingface.co/topabaem/LTX-2.5-Text-Encoder-4bit-8GB/resolve/main/samples/compare-det-02.mp4"></video>
 
 **Dune — the pair that does not.** Same terrain, same sun, same shadow, same
 walk, and a different person.
 
-<video controls width="100%" src="https://huggingface.co/topabaem/LTX-2.5-Text-Encoder-4bit/resolve/main/samples/compare-det-01.mp4"></video>
+<video controls width="100%" src="https://huggingface.co/topabaem/LTX-2.5-Text-Encoder-4bit-8GB/resolve/main/samples/compare-det-01.mp4"></video>
 
 Individually, the forge under the vendor's own sampler:
 
 | BF16 original | 4-bit |
 |---|---|
-| <video controls width="100%" src="https://huggingface.co/topabaem/LTX-2.5-Text-Encoder-4bit/resolve/main/samples/bf16-02.mp4"></video> | <video controls width="100%" src="https://huggingface.co/topabaem/LTX-2.5-Text-Encoder-4bit/resolve/main/samples/4bit-02.mp4"></video> |
+| <video controls width="100%" src="https://huggingface.co/topabaem/LTX-2.5-Text-Encoder-4bit-8GB/resolve/main/samples/bf16-02.mp4"></video> | <video controls width="100%" src="https://huggingface.co/topabaem/LTX-2.5-Text-Encoder-4bit-8GB/resolve/main/samples/4bit-02.mp4"></video> |
 
 `samples/README.md` carries the prompts, the per-clip conditioning drift, and
 what these clips do and do not establish. All thirty clips are in `samples/`,
@@ -178,13 +182,13 @@ Five packages and one file. No build step, no custom CUDA kernels, no
 compilation.
 
 ```bash
-pip install -r <(curl -sL https://huggingface.co/topabaem/LTX-2.5-Text-Encoder-4bit/resolve/main/requirements.txt)
+pip install -r <(curl -sL https://huggingface.co/topabaem/LTX-2.5-Text-Encoder-4bit-8GB/resolve/main/requirements.txt)
 ```
 
 ```python
 from huggingface_hub import hf_hub_download, snapshot_download
 
-repo = "topabaem/LTX-2.5-Text-Encoder-4bit"
+repo = "topabaem/LTX-2.5-Text-Encoder-4bit-8GB"
 
 # The loader ships with the weights; put it on the path before importing it.
 import sys, os
@@ -266,7 +270,7 @@ CONDITIONING.
 
 ## Try it
 
-**[Space: LTX-2.5 Text Encoder 4bit](https://huggingface.co/spaces/topabaem/LTX-2.5-Text-Encoder-4bit-Demo)**
+**[Space: LTX-2.5 Text Encoder 4bit](https://huggingface.co/spaces/topabaem/LTX-2.5-Text-Encoder-4bit-8GB-Demo)**
 — text-, image- and video-to-video, running this encoder on ZeroGPU. Measured
 there at 512x320, 25 frames: t2v 53.0 s, i2v 46.9 s, v2v 37.6 s.