v2.2.19
projectdiscovery/subfinderv2.2.19Sep 29, 2026by nocobase[bot]
Upgrading?
AI Summary
Adds support for OpenCode Zen/Go LLM services and DeepSeek Flash models with image attachments. Improves table column quick edit capabilities.
Key Highlights
- Added support for OpenCode Zen / Go as an LLM service
- DeepSeek Flash models now accept image attachments
- Table column quick edit: support setting a data scope
- Fix signing out and signing in with another account
- Prevent bulk update actions from submitting invalid option values
New Features
- OpenCode Zen / Go LLM service support
- Table column quick edit data scope support
Full Release Notes
### π New Features - **[AI employees]** Added support for OpenCode Zen / Go as an LLM service by sending the required session headers; DeepSeek Flash models now accept image attachments, and web search is limited to DeepSeek V4 Pro because Flash models silently ignore it ([#10551](https://github.com/nocobase/nocobase/pull/10551)) by @cgyrock ### π Improvements - **[client-v2]** Table column quick edit: support setting a data scope for association fields ([#10547](https://github.com/nocobase/nocobase/pull/10547)) by @katherinehhh ### π Bug Fixes - **[client-v2]** Fix the issue where signing in with another account after signing out shows a 404 page ([#10549](https://github.com/nocobase/nocobase/pull/10549)) by @zhangzhonghe - **[Action: Batch update]** Prevent bulk update actions from submitting option values that no longer exist in the configured collection field. ([#10540](https://github.com/nocobase/nocobase/pull/10540)) by @jiannx - **[Workflow]** Fix creation of the jobs status and ID index during database sync. ([#10552](https://github.com/nocobase/nocobase/pull/10552)) by @mytharcher
Related Repositories
ollama/ollamaβ
184,240
Get up and running with Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.
fatedier/frpβ
110,065
A fast reverse proxy to help you expose a local server behind a NAT or firewall to the internet.
infiniflow/ragflowβ
92,634
RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that fuses cutting-edge RAG with Agent capabilities to create a superior context layer for LLMs
gohugoio/hugoβ
90,197
The worldβs fastest framework for building websites.
junegunn/fzfβ
83,714
:cherry_blossom: A command-line fuzzy finder