Verified65% confidenceCautionExact time
Jev currently cannot explain the reasoning behind its judgments and does not support image input.
3
Sources
65%
Confidence
Medium-term (~90 days)
Relevance
9/24/2026
First Seen
Valid until: 12/23/2026
Sources
Related Entities
Related Claims
UnverifiedStarting with Jev 1.14, image input is accepted, and developers can put images into the state as file parts together with text, and ask Boolean or other types of questions about the image content.59% similarUnverifiedThere was one issue in the demonstration that did not fully work: screenshots could not be automatically attached to Jira.59% similarUnverifiedOpen-source image editing models Qwen Image Edit, FLUX.2 9B Base, and Hunyuan Image 3.0 Instruct all failed on POV generation tasks, unable to generate corresponding images from the perspective of a specific person in the image.57% similarUnverifiedv3.0.71 excludes screenshots and images from token estimation, to solve the problem of conversations with images prematurely triggering auto-compaction.57% similarUnverifiedThe creator did not use an image anchor, causing the characters in the video to undergo appearance changes between different seams.55% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/946982API
curl https://kongchang.com/api/v1/knowledge/claims/946982MCP
get_claim(id=946982)