openclaw - 💡(How to fix) Fix models.list is slow because it does provider startup and dependency installs on the hot path [2 comments, 2 participants]

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…
GitHub stats
openclaw/openclaw#74058Fetched 2026-04-30 06:29:15
View on GitHub
Comments
2
Participants
2
Timeline
4
Reactions
0
Author
Timeline (top)
commented ×2closed ×1cross-referenced ×1

Fix Action

Fix / Workaround

Additional detail from local reproduction:

  • openclaw gateway call models.list --json --timeout 20000 was fast once the gateway hot path was patched locally, which points at the gateway/catalog path rather than Ollama itself.
  • I also saw a separate QMD Node ABI mismatch in this environment, but that looked like a distinct issue rather than the main cause of models.list slowness.
RAW_BUFFERClick to expand / collapse

On OpenClaw 2026.4.23, models.list is taking far too long. In my environment it was often 45s to 2m+.

This does not appear to be Ollama itself: http://127.0.0.1:11434/api/tags responded quickly. The slowdown came from OpenClaw doing extra work during model catalog load, including provider plugin loading, bundled runtime dependency installs, and pricing fetches that timed out.

Observed:

  • gateway log showed models.list 45136ms
  • CLI openclaw models list --json took about 1:10 to 2:11
  • logs showed repeated npm install ... work during model listing/startup
  • remote pricing fetches timed out at 30s

Expected: models.list should return quickly from cached or already-resolved catalog data, without running heavy installs or slow network fetches on the request path.

Additional detail from local reproduction:

  • openclaw gateway call models.list --json --timeout 20000 was fast once the gateway hot path was patched locally, which points at the gateway/catalog path rather than Ollama itself.
  • I also saw a separate QMD Node ABI mismatch in this environment, but that looked like a distinct issue rather than the main cause of models.list slowness.

extent analysis

TL;DR

The slowdown in models.list can be mitigated by optimizing the model catalog load process, potentially by caching or pre-loading dependencies and avoiding slow network fetches.

Guidance

  • Investigate the possibility of caching or pre-loading provider plugins and bundled runtime dependencies to avoid repeated npm install work during model listing.
  • Consider optimizing or disabling remote pricing fetches that are timing out at 30s, or increasing the timeout value if necessary.
  • Review the gateway/catalog path for potential bottlenecks or areas for improvement, as the local patch to the gateway hot path resulted in faster models.list responses.
  • Verify that the QMD Node ABI mismatch is resolved, as it may be contributing to the issue or causing other problems.

Example

No code snippet is provided as the issue does not imply a specific code change.

Notes

The root cause of the slowdown appears to be related to the model catalog load process, but further investigation is needed to determine the best solution. The QMD Node ABI mismatch may be a separate issue that needs to be addressed.

Recommendation

Apply a workaround to optimize the model catalog load process, such as caching or pre-loading dependencies, to improve the performance of models.list. This is recommended as it directly addresses the observed slowdown and can potentially provide a significant improvement in response time.

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…

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 models.list is slow because it does provider startup and dependency installs on the hot path [2 comments, 2 participants]