Qwen Code adds a separate image-generation model route with PNG safety checks
qqqys · ghdev · 2026-07-23
Qwen Code adds a configurable image-generation route
This PR adds a user-selectable image-only model to Qwen Code, alongside the existing voice and vision auxiliary models. Users can pick it with /model --image, and the CLI invokes an approval-gated image-generation tool that writes verified PNGs as workspace artifacts.
Key implementation details:
- image models are kept separate from the primary chat model and excluded from chat/fast/voice/vision/fallback/subagent selection paths
- the initial transport uses Qwen’s synchronous multimodal image-generation request/response format
- downloads are locked down with HTTPS-only policy, DNS pinning across redirects, bounded response sizes, PNG signature checks, timeout/cancellation handling, redacted errors, and atomic workspace writes
The motivation is to avoid conflating chat and generation protocols while still letting users bring their own endpoint and credentials.
More from coding & agent
- Reticle builds a verification layer that lets AI coding agents check their own generated code — Scobleizer · 2026-10-27
- Trust your agent, but read its logs: Claude managed agents tips — brada · 2026-10-27
- Dev training a guard model to block prompt injection, tool and memory poisoning — gnukeith · 2026-10-27
- Dev trains a ~4GB-VRAM sentinel model to guard against prompt injection and poisoned memory — gnukeith · 2026-10-27
- Superset rebrands as Paradev, raises $11.5M seed led by USV — massimosgrelli · 2026-10-27
- Ex-Big Tech managers may thrive in the AI era: manage agents like you managed teams — lxfater · 2026-10-27