openclaw - ✅(Solved) Fix Bug: GLM provider does not register media-understanding capability despite image input config [1 pull requests, 1 comments, 2 participants]

openclaw2026-03-21 03:03:52

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.

GitHub issue URL

Helpful · Quick feedback

GitHub stats

openclaw/openclaw#51392•Fetched 2026-04-08 01:11:49

View on GitHub

Comments

Participants

Timeline

Reactions

Author

cpc1986

Participants

cpc1986

Ryce

Timeline (top)

referenced ×3cross-referenced ×2closed ×1commented ×1

Error Message

Error: "No media-understanding provider registered for glm"

Code Example

{
  "models": {
    "providers": {
      "glm": {
        "models": [{
          "id": "glm-4.6v",
          "input": ["text", "image"]
        }]
      }
    }
  },
  "agents": {
    "defaults": {
      "imageModel": { "primary": "glm/glm-4.6v" }
    }
  }
}

RAW_BUFFERClick to expand / collapse

Bug Description

When a GLM model (e.g. glm-4.6v) is configured with input: ["text", "image"], the OpenClaw provider does not automatically register the media-understanding capability. This causes the image tool to fail with:

No media-understanding provider registered for glm

Reproduction

Configure a GLM model with input: ["text", "image"] as agents.defaults.imageModel
Call the image tool with any image
Error: "No media-understanding provider registered for glm"

Expected Behavior

When a model's input array contains "image", the provider should auto-register the media-understanding capability, enabling the image tool.

Config

{
  "models": {
    "providers": {
      "glm": {
        "models": [{
          "id": "glm-4.6v",
          "input": ["text", "image"]
        }]
      }
    }
  },
  "agents": {
    "defaults": {
      "imageModel": { "primary": "glm/glm-4.6v" }
    }
  }
}

extent analysis

Fix Plan

To fix the issue, we need to auto-register the media-understanding capability when a GLM model is configured with input: ["text", "image"].

Code Changes

We can achieve this by modifying the OpenClaw provider to check the model's input configuration and register the capability if necessary. Here's an example code snippet:

# openclaw_provider.py

def register_capabilities(self, model):
    if "image" in model["input"]:
        self.register_capability("media-understanding")

# In the model configuration loop
for model in models:
    # ... existing code ...
    register_capabilities(model)

Alternatively, we can also modify the image tool to check if the media-understanding capability is registered before attempting to use it:

# image_tool.py

def __init__(self, model):
    if "image" in model["input"] and not self.is_capability_registered("media-understanding"):
        raise ValueError("Media-understanding capability not registered for model")

Configuration Changes

No configuration changes are required. The fix relies solely on code modifications.

Verification

To verify that the fix worked, follow these steps:

Configure a GLM model with input: ["text", "image"] as agents.defaults.imageModel
Call the image tool with any image
The image tool should no longer raise the "No media-understanding provider registered for glm" error

Extra Tips

Make sure to test the fix with different model configurations to ensure that the media-understanding capability is registered correctly in all cases.
Consider adding additional logging or debugging statements to help diagnose any future issues related to capability registration.

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?

#api #ssr #installation #tensor shape #autograd error #model save/load #optimization #mixed precision #training loop #device allocation

Still need to ship something?

×6

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

openclaw - ✅(Solved) Fix Bug: GLM provider does not register media-understanding capability despite image input config [1 pull requests, 1 comments, 2 participants]

Recommended Tools

GitHub issue graph ai analysis

Error Message

Fix Action

Fixed

PR fix notes

PR #51418: fix(media-understanding): auto-register image capability

Description (problem / solution / changelog)

Summary

Change Type (select all)

Scope (select all touched areas)

Linked Issue/PR

User-visible / Behavior Changes

Security Impact (required)

Repro + Verification

Environment

Steps

Expected

Actual

Evidence

Human Verification (required)

Review Conversations

Compatibility / Migration

Failure Recovery (if this breaks)

Risks and Mitigations

Changed files