Unverified50% confidenceReleaseExact time
Starting with OpenHands v1.27.0, the model list is dynamically fetched and cached by the local backend spinning up an agent server, instead of using the hardcoded static list in Canvas.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/23/2026
First Seen
Valid until: 1/21/2027
Sources
Related Entities
Related Claims
UnverifiedOpenHands' Agent Canvas can run OpenHands, Claude Code, Codex, Gemini CLI or OpenCode on a laptop, in Docker or on your own server.66% similarUnverifiedOpenHands v1.27.0 adds OpenCode and PI to the built-in agent list, allowing users to run these two coding agents directly as OpenHands backends.64% similarUnverifiedThe OpenAI Agents SDK registers tools inline directly, eliminating the need to manually set up an MCP Server, and the model field is required.64% similarUnverifiedOpenHands v1.25.0 (October 6) released the model router, voice dictation and configurable workspace file discovery; v1.26.0 (October 8) released backend suspension explanations with one-click switching and officially verified skills; v1.27.0 (October 12) released OpenCode/PI built-in support and model list reporting; v1.27.1 was a hotfix; v1.28.0 released in-conversation secret requests and streaming fade-in output.61% similarUnverifiedThis article, based on openai-agents version 0.23.1, verified three integration approaches against a local mock service that implements only /v1/chat/completions.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1091916API
curl https://kongchang.com/api/v1/knowledge/claims/1091916MCP
get_claim(id=1091916)