[GH-ISSUE #6229] ollama运行多模态大模型如llava时,不支持多张图片输入 - support multiple image inputs to llava #50405

Closed
opened 2026-04-28 15:38:33 -05:00 by GiteaMirror · 2 comments
Owner

Originally created by @lyp-liu on GitHub (Aug 7, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/6229

api参考文档写了可以发送文档,为什么用llava等多模态大模型时只能识别到列表最后一张图片?或者需要修改modelfile中的template吗

Originally created by @lyp-liu on GitHub (Aug 7, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/6229 api参考文档写了可以发送文档,为什么用llava等多模态大模型时只能识别到列表最后一张图片?或者需要修改modelfile中的template吗
GiteaMirror added the needs more infofeature request labels 2026-04-28 15:38:33 -05:00
Author
Owner

@dhiltgen commented on GitHub (Nov 6, 2024):

llava sometimes doesn't handle multiple images well. Please try ollama version 0.4.0 and llava:34b

<!-- gh-comment-id:2458486944 --> @dhiltgen commented on GitHub (Nov 6, 2024): llava sometimes doesn't handle multiple images well. Please try ollama version 0.4.0 and llava:34b
Author
Owner

@pdevine commented on GitHub (Mar 21, 2025):

Going to go ahead and close this since there hasn't been an update.

<!-- gh-comment-id:2744182015 --> @pdevine commented on GitHub (Mar 21, 2025): Going to go ahead and close this since there hasn't been an update.
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#50405