Document 4096-token release defaults

README.md

@@ -27,9 +27,10 @@ llama.cpp WebAssembly/WebGPU.
 - **Bonsai 27B** — WebGPU-only hybrid model with strict adapter and full-graph
   checks; the UI refuses silent CPU fallback.
 
-The chat UI allocates a 4,000-token context and allows up to 2,000 completion
-tokens by default. Prompt text, system instructions, tool schemas, and prior
-turns share the same context window with the completion.
+The chat UI allocates a 4,096-token context and requests up to 4,096 completion
+tokens by default. Prompt text, system instructions, tool schemas, prior turns,
+and generated tokens share that same context window, so 4,096 is a completion
+ceiling rather than a guarantee after a non-empty prompt.
 
 The Space manifest pins
 [`WaveCut/Bonsai-web-GGUF`](https://huggingface.co/WaveCut/Bonsai-web-GGUF)