On October 24, 2026, at 12:47 Brasília time, Qwen Code published version v0.26.1, marked as the latest in the official repository. The release updates the CLI and managed sessions, without announcing a new model.
Qwen Code 0.26.1: save the daemon token with --save-token
Qwen Code 0.26.1 adds qwen daemon start --save-token, context compression for idle sessions and session isolation fixes. Here is how to use them.
Marcelo Santiago
Criador do bridgecode
The biggest AI newsletter in the World
Get AI articles in your inbox
Technical curation on models, agents and coding tools. No spam, unsubscribe anytime.
In this article you will see:
- How optional token storage works
- What changes in context compression
- What adjustments arrived to session isolation
Qwen Code 0.26.1 makes daemon token storage optional
With qwen daemon start --save-token, the daemon can save the bearer token in a file accessible only by the owner. The command qwen daemon token manages this token, and qwen daemon stop removes it.
Credit and source: GitHub.
Qwen Code 0.26.1 compresses context of idle sessions
The release adds manual context compression only when managed sessions linked to a workspace are idle. The feature is exposed via the public API and in the WebShell context usage indicator, preserving the audit history.
The version also restores QQBot session isolation by group in thread mode. To compare with a recent Qwen model test, see the ThinkingBox benchmark with Qwen3.8-27B.
Frequently asked questions about Qwen Code 0.26.1
What does Qwen Code 0.26.1 add to the daemon?
The command `qwen daemon start --save-token` can save the bearer token to a file accessible only by the owner; `qwen daemon token` manages it and `qwen daemon stop` removes it.
Does Qwen Code 0.26.1 compress context automatically?
No. Compression is manual and is only available when managed sessions linked to a workspace are idle.
Which sessions get manual context compression?
The changelog mentions managed sessions linked to a workspace, accessible via the public API and the context usage indicator in WebShell.
Is Qwen Code 0.26.1 a new Qwen model?
No. The release updates the CLI and managed sessions, without announcing new model weights.
Qwen Code 0.26.1 updates tools, not model weights
The changelog describes changes to the CLI and managed sessions, not the launch of a new model. Anyone evaluating open options for coding can check the list of open source code assistants.
The practical point is that persistent token is opt-in, while context compression is manual and limited to idle sessions.
You set the goal and the agents split the tasks: one plans, another executes, another reviews. See how orchestration works on bridgecode.
Source: GitHub
Tags:#Qwen#Qwen Code#release#daemon#tokens#Managed Sessions#agentes de código
Developer and creator of bridgecode, an agentic development environment for orchestrating squads of AI agents. Helps people going from vibe coding to professional development move from idea to shipped product.
Related articles
Newsletter
Karpathy builds a video platform over a weekend with Claude
Andrej Karpathy built HackerTube in a weekend with Claude. What it says about development speed with coding agents.
5 min
Newsletter
Rich Sutton reinforces that compute alone does not create intelligence
Turing Award winner Rich Sutton reinforces that raw compute is not enough for intelligence and calls for new methods, continuous learning and world models.
5 min
Newsletter
AI auditors detect misbehavior but generate false alarms
AI models catch a good portion of other agents' misbehavior, but they generate a lot of false alarms. See the numbers from Andy Hall's study.
6 min