@@ -49,8 +49,9 @@ const OUTPUT_TOKEN_FALLBACK = 32_000
 // No reply needs more, however much the model allows.
 const OUTPUT_TOKEN_MAX = 256_000
 // A title never needs more; without a bound a local thinking model can reason in circles
-// for minutes on the background title request while the user waits (#54401).
-const TITLE_OUTPUT_MAX = 512
+// for minutes on the background title request while the user waits (#54401). Sized to leave
+// room for a small thinking budget, so budget-aware providers don't reject the request.
+const TITLE_OUTPUT_MAX = 4_096
 // Prompt text is estimated at about 4 characters per token, which can run low on dense text such as code.
 const ESTIMATE_ERROR = 0.15
 // Never ask for less; only reachable with automatic compaction off, since it keeps the window from filling this far.
 
@@ -688,6 +688,6 @@ it.effect("bounds the title request with a small output limit (#54401)", () =>
     expect(requests).toHaveLength(1)
     const maxTokens = requests[0]?.generation?.maxTokens
     expect(maxTokens).toBeDefined()
-    expect(maxTokens!).toBeLessThanOrEqual(512)
+    expect(maxTokens!).toBeLessThanOrEqual(4_096)
   }),
 )
Original file line number	Diff line number	Diff line change
`@@ -688,6 +688,6 @@ it.effect("bounds the title request with a small output limit (#54401)", () =>`
`688`	`688`	`expect(requests).toHaveLength(1)`
`689`	`689`	`const maxTokens = requests[0]?.generation?.maxTokens`
`690`	`690`	`expect(maxTokens).toBeDefined()`
`691`		`- expect(maxTokens!).toBeLessThanOrEqual(512)`
	`691`	`+ expect(maxTokens!).toBeLessThanOrEqual(4_096)`
`692`	`692`	`}),`
`693`	`693`	`)`