Claude Code v2.1.304 · 19 Oct 2026
Bundled default model changes to Opus 5.5 at medium effort
The model catalog shipped with Claude Code now defaults to claude-opus-5-5 at medium effort instead of claude-opus-5 at high effort
You'll notice Tier: how much it should matter to you
5 Useful: my rating, 1 to 5
4 Signal: worth watching, 1 to 5
Models Area: what it touches
Improvements Kind: in v2.1.304, 19 Oct 26
What probably matters to you Section of the release
What
Claude Code ships with a signed model catalog (a list of models and settings it trusts because it is cryptographically signed). The catalog's default entry for Claude Code has changed:
Thinking effort is how much reasoning the model does before answering.
The catalog can be refreshed from the server while Claude Code runs, so the copy shipped with the app can be replaced.
Why
If you rely on the default, the model you get and how much it thinks before answering can change, with medium effort as the new default.
See this entry in the whole of v2.1.304 →
19 Oct 2026 · 371 entries, this one is #9.
20 186 7 158
Open the release →
Names in this entry
model claude-opus-5-5 New Opus 5.5 model added with a 1M-token context window and its own pricing tier
model claude-opus-5 Opus 5 is now the default Opus, with a 1M-context option in the model picker.
Is this right?
0 0 Feedback
What's wrong with this entry?
Hard to understand Too vague Looks wrong Not relevant to me Too long Missing the detail I wanted Duplicate of another card
Clear Learned something new I can act on this Good catch
Related
Other releases about the same thing. Found by shared names or similar wording; neither means one caused the other.
← Previous · 8 of 371
New CLAUDE_CODE_ATTRIBUTION_PROJECT setting checked against claude.ai Project IDs
Use it now Projects
In v2.1.304 What probably matters to you release-page order
Next → · 10 of 371
SDK hosts can mark MCP tools that run code, so auto mode checks them
Use it now SDK
