From bf3763526ab5f08f5ef1fe5598c084189ef56ac8 Mon Sep 17 00:00:00 2001
From: b1tg <33436708+b1tg@users.noreply.github.com>
Date: Wed, 8 Apr 2026 10:26:23 +0800
Subject: [PATCH] llm: buffer SSE chunks to fix parse errors from split reads
 (#15641)

---
 tinygrad/apps/llm.py | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/tinygrad/apps/llm.py b/tinygrad/apps/llm.py
index 6fe4a19201..9ffa500d4f 100644
--- a/tinygrad/apps/llm.py
+++ b/tinygrad/apps/llm.py
@@ -313,10 +313,14 @@ CHAT_HTML = b'''<!DOCTYPE html><html><head><title>tinygrad chat</title><style>
     const d = document.createElement('div'); d.className = 'msg'; chat.appendChild(d);
     const r = await fetch('/v1/chat/completions', {method: 'POST', headers: {'Content-Type': 'application/json'},
       body: JSON.stringify({model: 'llama', messages: msgs, stream: true, temperature: 0.7})});
+    let buf = '';
     for (const rd = r.body.getReader(), dec = new TextDecoder();;) {
       const {done, value} = await rd.read();
       if (done) break;
-      for (const ln of dec.decode(value).split('\\n'))
+      buf += dec.decode(value, {stream: true});
+      const lines = buf.split('\\n');
+      buf = lines.pop();
+      for (const ln of lines)
         if (ln.startsWith('data: ') && !ln.includes('[DONE]'))
           try { d.textContent += JSON.parse(ln.slice(6)).choices[0]?.delta?.content || '' } catch {}
       chat.scrollTop = chat.scrollHeight;
