Version: 1.124.2 (Universal)
Commit: 6928394
Date: 2026-06-12T00:12:36Z (4 days ago)
Electron: 42.2.0
ElectronBuildId: 14159160
Chromium: 148.0.7778.97
Node.js: 24.15.0
V8: 14.8.178.14-electron.0
OS: Darwin arm64 25.1.0
When selecting Claude Sonnet 4.6 in the chat model picker with the context size set to 200K (the Default option), the actual request sent to the model uses the 1M context window instead.
Steps to reproduce:
- Open the Chat view and select Claude Sonnet 4.6 as the model.
- In the context size dropdown, select 200K (marked as Default).
- Send a message.
- Open the request log and inspect the request metadata.
Expected behavior:
With the 200K context size selected, maxPromptTokens + maxResponseTokens should reflect a ~200K total context window (e.g. maxPromptTokens of roughly 136000 with maxResponseTokens of 64000).
Actual behavior:
The request metadata shows a 1M-sized context window:
model: claude-sonnet-4.6
maxPromptTokens: 936000
maxResponseTokens: 64000
This corresponds to the 1M option, not the selected 200K

Version: 1.124.2 (Universal)
Commit: 6928394
Date: 2026-06-12T00:12:36Z (4 days ago)
Electron: 42.2.0
ElectronBuildId: 14159160
Chromium: 148.0.7778.97
Node.js: 24.15.0
V8: 14.8.178.14-electron.0
OS: Darwin arm64 25.1.0
When selecting Claude Sonnet 4.6 in the chat model picker with the context size set to 200K (the Default option), the actual request sent to the model uses the 1M context window instead.
Steps to reproduce:
Expected behavior:
With the 200K context size selected,
maxPromptTokens+maxResponseTokensshould reflect a ~200K total context window (e.g.maxPromptTokensof roughly 136000 withmaxResponseTokensof 64000).Actual behavior:
The request metadata shows a 1M-sized context window:
model:claude-sonnet-4.6maxPromptTokens:936000maxResponseTokens:64000This corresponds to the 1M option, not the selected 200K