From b9854716378c39a36bf751dc5cb36f6456e697b3 Mon Sep 17 00:00:00 2001
From: Daniel Han <danielhanchen@gmail.com>
Date: Mon, 16 Mar 2026 07:04:07 +0000
Subject: [PATCH] Increase default max tokens to 8192, disable repetition
 penalty

- maxTokens: 2048 -> 8192. The old 2048 limit caused generation to
  stop mid-output for longer responses (e.g. reasoning/thinking models
  that produce long chain-of-thought before the answer).
- repetitionPenalty: 1.1 -> 1.0 (disabled). Most models handle
  repetition well on their own. A penalty of 1.1 can hurt quality
  for creative tasks like code generation and ASCII art.
- Change welcome message from "Run LLMs or test your fine-tune" to
  "Chat with your model".
---
 studio/frontend/src/components/assistant-ui/thread.tsx | 2 +-
 studio/frontend/src/features/chat/types/runtime.ts     | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/studio/frontend/src/components/assistant-ui/thread.tsx b/studio/frontend/src/components/assistant-ui/thread.tsx
index 2a1e309782..0111c2085a 100644
--- a/studio/frontend/src/components/assistant-ui/thread.tsx
+++ b/studio/frontend/src/components/assistant-ui/thread.tsx
@@ -139,7 +139,7 @@ const ThreadWelcome: FC<{ hideComposer?: boolean }> = ({ hideComposer }) => {
               className="size-20"
             />
             <h1 className="aui-thread-welcome-message-inner fade-in slide-in-from-bottom-1 animate-in font-semibold text-2xl duration-200">
-              Run LLMs or test your fine-tune
+              Chat with your model
             </h1>
             <p className="aui-thread-welcome-message-inner fade-in slide-in-from-bottom-1 animate-in text-muted-foreground text-base delay-75 duration-200">
               Run GGUFs, safetensors, vision and audio models!
diff --git a/studio/frontend/src/features/chat/types/runtime.ts b/studio/frontend/src/features/chat/types/runtime.ts
index 3a5f803de5..a40ec7ceb8 100644
--- a/studio/frontend/src/features/chat/types/runtime.ts
+++ b/studio/frontend/src/features/chat/types/runtime.ts
@@ -20,9 +20,9 @@ export const DEFAULT_INFERENCE_PARAMS: InferenceParams = {
   topP: 0.9,
   topK: 50,
   minP: 0.01,
-  repetitionPenalty: 1.1,
+  repetitionPenalty: 1.0,
   maxSeqLength: 4096,
-  maxTokens: 2048,
+  maxTokens: 8192,
   systemPrompt: "",
   checkpoint: "",
   trustRemoteCode: false,