README: note input embedding is loaded from verifier (not shipped), checkpoint ~3.7GB

README.md

@@ -35,6 +35,9 @@ Compared with an MHA draft model, the MLA variant is a better fit for Kimi-K2.7-
 - **Algorithm**: EAGLE-3 with MLA, single draft decoder layer.
 - **Verifier**: Kimi-K2.7-Code. The draft reuses the verifier's frozen embedding / lm_head / norm
   and trains one MLA decoder layer plus an auxiliary-hidden-state fusion layer.
+- **Shared input embedding**: the input embedding is **not shipped** in this checkpoint — it is
+  loaded from the verifier at inference (vLLM's `Eagle3DeepseekV2ForCausalLM` shares the target's
+  `embed_tokens` when the draft weights omit it). This keeps the checkpoint compact (~3.7 GB).
 - **Draft vocabulary**: full 163,840-token vocabulary (no truncation).
 
 ### Training Setup