Issue Station
Sign In
Tools hereToolbox

openclaw - 💡(How to fix) Fix imageModel not resolved for agents with per-agent model override

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…

Root Cause

When an agent has a model override (e.g. ollama/glm-5.1:cloud) but no per-agent imageModel (which is impossible because the schema has additionalProperties: false and doesn't include imageModel as a valid key on agents.list.*), the image tool fails to resolve a vision model.

Fix Action

Fix / Workaround

Workaround attempted: setting vision-capable fallbacks on the agent model — resolver still doesn't pick them up for the image tool.

RAW_BUFFERClick to expand / collapse

When an agent has a model override (e.g. ollama/glm-5.1:cloud) but no per-agent imageModel (which is impossible because the schema has additionalProperties: false and doesn't include imageModel as a valid key on agents.list.*), the image tool fails to resolve a vision model.

The agents.defaults.imageModel config IS set (e.g. ollama/kimi-k2.5:cloud) but the resolver doesn't fall back to it when the agent's primary model is text-only.

The per-agent schema (agents.list.*) has additionalProperties: false and no imageModel key, so there's no way to set a per-agent image model override.

Agents without model overrides work fine — they inherit the default imageModel.

Workaround attempted: setting vision-capable fallbacks on the agent model — resolver still doesn't pick them up for the image tool.

OpenClaw version: 2026.4.15

extent analysis

TL;DR

The image tool fails to resolve a vision model when an agent has a model override without a per-agent imageModel, and the resolver doesn't fall back to the default imageModel.

Guidance

  • The issue seems to be related to the resolver not falling back to the default imageModel when the agent's primary model is text-only, suggesting a potential logic flaw in the resolver.
  • Verify that the agents.defaults.imageModel config is correctly set and that the resolver is properly configured to use this default when no per-agent imageModel is provided.
  • Consider modifying the resolver to explicitly check for the presence of a per-agent imageModel and fall back to the default imageModel if it's not present.
  • Check the OpenClaw version documentation (2026.4.15) to see if there are any known issues or configuration options related to model resolution and fallback behavior.

Example

No code snippet is provided as the issue seems to be related to configuration and resolver logic rather than code.

Notes

The issue is specific to the OpenClaw version 2026.4.15, and the solution may vary depending on the specific configuration and requirements of the system.

Recommendation

Apply a workaround by modifying the resolver to fall back to the default imageModel when no per-agent imageModel is provided, as the current behavior seems to be a logic flaw rather than a version-specific issue.

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…
#model download#tokenizer error#prompt formatting#chain error#conversation history

Still need to ship something?

×6

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

Back to top recommendations

TRENDING

openclaw - 💡(How to fix) Fix imageModel not resolved for agents with per-agent model override