Issue Station
Sign In
Tools hereToolbox

hermes - 💡(How to fix) Fix WeChat gateway cannot process images with Xiaomi MiMo models [1 pull requests]

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…

When using Xiaomi MiMo models (e.g., mimo-v2-5) as the default model with the WeChat gateway, images sent by users cannot be processed. The vision_analyze tool returns an error, and images in conversations are not recognized by the model.

However, the same model works perfectly for image understanding in CLI mode (hermes chat).

Error Message

Error code: 400 - {'error': {'message': "Failed to deserialize the JSON body into the target type: messages[0]: unknown variant image_url, expected text", 'type': 'invalid_request_error', 'param': None, 'code': 'invalid_request_error'}}

Root Cause

When using Xiaomi MiMo models (e.g., mimo-v2-5) as the default model with the WeChat gateway, images sent by users cannot be processed. The vision_analyze tool returns an error, and images in conversations are not recognized by the model.

However, the same model works perfectly for image understanding in CLI mode (hermes chat).

Fix Action

Fixed

RAW_BUFFERClick to expand / collapse

Description

When using Xiaomi MiMo models (e.g., mimo-v2-5) as the default model with the WeChat gateway, images sent by users cannot be processed. The vision_analyze tool returns an error, and images in conversations are not recognized by the model.

However, the same model works perfectly for image understanding in CLI mode (hermes chat).

Steps to Reproduce

  1. Set default model to mimo-v2-5 with xiaomi provider
  2. Send an image via WeChat (Weixin)
  3. The image is downloaded to cache (e.g., ~/.hermes/image_cache/img_xxx.jpg)
  4. Model responds that it cannot see the image, or vision_analyze tool fails

Error

Error code: 400 - {'error': {'message': "Failed to deserialize the JSON body into the target type: messages[0]: unknown variant image_url, expected text", 'type': 'invalid_request_error', 'param': None, 'code': 'invalid_request_error'}}

Expected Behavior

Images sent via WeChat should be processed by multimodal models (MiMo-V2.5, MiMo-V2-Omni) just like they are in CLI mode.

Environment

  • Hermes Agent latest
  • Gateway: Weixin (iLink Bot API)
  • Model: mimo-v2-5 (xiaomi provider)
  • Platform: WSL2 on Windows

Notes

  • CLI mode works fine with the same model for image understanding
  • The issue appears to be in how the gateway formats image data before sending to the MiMo API endpoint

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…
#api#container setup#orchestration issue#cache issue#memory leak

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING