Correct the VRAM figure: allocated bytes undercount the card by 2 GiB

README.md

@@ -139,16 +139,36 @@ Individually, the forge under the vendor's own sampler:
 what these clips do and do not establish. All thirty clips are in `samples/`,
 and the Space plays them side by side under its **BF16 vs 4-bit** tab.
 
-## Memory
+## Memory — what the card actually needs
 
-| mode | resident | encode |
-|---|---:|---|
-| `resident` (default) | **7.75 GiB** | dequantizes inside `forward` |
-| `dequantized` | ~26.3 GB | one dense build at load |
-
-The two produce `torch.equal` conditioning, so the choice is footprint against
-speed and never quality. `resident` measured at **8.34 GiB** peak on a 16 GB
-V100 — the encoder fits one consumer card for the first time.
+| | resident (default) | dequantized |
+|---|---:|---:|
+| PyTorch allocated, peak | 8.48 GiB | ~22.3 GiB |
+| PyTorch reserved, peak | 9.33 GiB | — |
+| **what `nvidia-smi` shows** | **9.70 GiB** | — |
+
+**Size your card from the last row.** The first is `max_memory_allocated`, which
+counts only live allocator blocks — it misses the CUDA context and everything
+the caching allocator reserved and has not handed back, and it undercounts by
+nearly 2 GiB here. Earlier versions of this card quoted that number, and anyone
+who bought an 8 GB card on the strength of it would have been wrong.
+
+Measured on a **Tesla V100-SXM2-16GB (cc 7.0)** with a 55-token prompt, through
+the ComfyUI-faithful path that left-pads to 1024 tokens. So:
+
+* **16 GB and up** — comfortable.
+* **12 GB** — fits.
+* **10 GB** — fits, with roughly 300 MB of headroom. Nothing else on the card.
+* **8 GB** — does not fit.
+
+The two modes produce `torch.equal` conditioning, so the choice is footprint
+against speed and never quality. Resident dequantizes inside `forward`, which
+costs time on short prompts; dequantized builds one dense BF16 model at load and
+then needs a card that can hold 26 GB.
+
+These figures are for the **text encoder alone**. Generating video also needs the
+DiT and the VAEs, which this repository does not contain — the pipeline in the
+Space loads a `Q3_K_M` DiT alongside it.
 
 ## Use it