Model card: add Serving with vLLM section

Documents that vLLM serves this checkpoint natively, that the chat template now
supplies the latent block automatically, how to check the latent path is
actually active (8.9 vs 101.2 mean output tokens with/without), and the
max_pixels setting needed to reproduce the paper's numbers.

README.md

@@ -115,6 +115,57 @@ response = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]
 print(response)
 ```
 
+## Serving with vLLM
+
+vLLM serves this checkpoint with its **native** Qwen2.5-VL implementation — no
+custom model code and no `--trust-remote-code`. The latent tokens are ordinary
+entries in an extended vocabulary (`vocab_size` 151669) with trained embeddings,
+and the latent-specific code paths in the training model are gated on a
+training-only flag, so they never run during generation.
+
+```bash
+vllm serve array/Qwen2.5-VL-Mull --max-model-len 32768
+```
+
+The chat template pre-fills the assistant turn with `<think>` + 20 x
+`<|latent_pad|>` + `</think>`, so ordinary chat requests get the latent block
+automatically:
+
+```python
+from openai import OpenAI
+client = OpenAI(base_url="http://localhost:8000/v1", api_key="EMPTY")
+client.chat.completions.create(
+    model="array/Qwen2.5-VL-Mull",
+    messages=[{"role": "user", "content": [
+        {"type": "image_url", "image_url": {"url": f"data:image/jpeg;base64,{b64}"}},
+        {"type": "text", "text": question},
+    ]}],
+    temperature=0, max_tokens=512,
+    # optional: vary the latent budget, or set 0 to disable it for an ablation
+    extra_body={"chat_template_kwargs": {"num_latents": 20}},
+)
+```
+
+**Check that the latent path is active.** If you serve this model with a stock
+Qwen2.5-VL chat template, the prompt ends at `<|im_start|>assistant\n` and the
+model receives *zero* latent tokens. It still answers, so the mistake is easy to
+miss — but it reverts to writing its reasoning out as text. On 24 SAT items the
+same checkpoint averaged **8.9 output tokens** with the latent block and
+**101.2** without. Verify with
+`list(output.prompt_token_ids).count(151665) == 20`.
+
+**Reproducing the paper's numbers.** The evaluations run
+`max_pixels=12845056`, while `preprocessor_config.json` defaults to 401408:
+
+```bash
+vllm serve array/Qwen2.5-VL-Mull --max-model-len 32768 \
+    --mm-processor-kwargs '{"min_pixels": 3136, "max_pixels": 12845056}'
+```
+
+A [recipe](https://recipes.vllm.ai/array/Qwen2.5-VL-Mull), serving scripts and an
+HF-vs-vLLM parity harness live in
+[`serving/`](https://github.com/arijitray1993/mull/tree/main/serving).
+
 ## Citation
 
 ```bibtex