Guide

OpenClaw v2026.3.28: Google/models, recovery notes

March 29, 20264 min readUpdated September 11, 2026By OpenClawBlog Team

v2026.3.28 is easiest to understand through failure. The official record describes a provider, model, voice, or context change; the sections below spell out what to save, what to interrupt, and what must remain available afterward.

Stable status describes the release line, not every local provider or channel. Check the version-specific notes before upgrading. Treat the version as one input to the decision. Local credentials, enabled plugins, channels, and host support still need their own evidence.

Channel
Stable

Primary signals
xAI/tools; xAI/onboarding; MiniMax

Publication date
2026-03-29

Failure and recovery path

xAI/tools

Evidence from the release. move the bundled xAI provider to the Responses API, add first-class x_search, and auto-enable the xAI plugin from owned web-search and tool config so bundled Grok auth/configured search flows work without manual plugin toggles.

Operator reading. Before widening access for xAI/tools, replay a narrow allow and deny case. A readable denial and a revoked test credential are part of the acceptance record.

xAI/onboarding

Evidence from the release. let the bundled Grok web-search plugin offer optional x_search setup during openclaw onboard and openclaw configure --section web, including an x_search model picker with the shared xAI key.

Operator reading. The safe reading of xAI/onboarding is operational: prove the model can answer the intended task, then force a bounded failure and verify that no side effect is repeated.

MiniMax

Evidence from the release. add image generation provider for image-01 model, supporting generate and image-to-image editing with aspect ratio control.

Operator reading. The safe reading of MiniMax is operational: prove the model can answer the intended task, then force a bounded failure and verify that no side effect is repeated.

Agents/Anthropic

Evidence from the release. recover unhandled provider stop reasons (e.g. sensitive) as structured assistant errors instead of crashing the agent run.

Operator reading. Check Agents/Anthropic with a known prompt and a disposable account. Save the selected route, reasoning setting, latency, and refusal or fallback result so another operator can reproduce it.

The limits of the recovery promise

Google/models

resolve Gemini 3.1 pro, flash, and flash-lite for all Google provider aliases by passing the actual runtime provider ID and adding a template-provider fallback; fix flash-lite prefix ordering.

Check Google/models with a known prompt and a disposable account. Save the selected route, reasoning setting, latency, and refusal or fallback result so another operator can reproduce it.

Run the smallest useful rehearsal

Write down the baseline for xAI/tools, including the installed version, host, provider, and workspace.
Confirm what can be restored, then take a verified backup before applying the candidate package.
Use a disposable session to compare the new model or provider with the previous route on one fixed input.
Restore the previous provider setting after the check and retain the measured result beside the release tag.
Keep the experiment bounded; an unexpected result belongs in the release record before any wider rollout.

openclaw --version
openclaw gateway status
openclaw security audit

Leave a usable maintenance note

Save the command, output, source link, and environment used for this check. Future operators should be able to tell whether a difference belongs to the release or to local provider, channel, or platform state.

Fix details to carry forward

Agents/Anthropic. recover unhandled provider stop reasons (e.g. sensitive) as structured assistant errors instead of crashing the agent run. This is the edge case to include in a regression check, especially when the release is pre-release.

Google/models. resolve Gemini 3.1 pro, flash, and flash-lite for all Google provider aliases by passing the actual runtime provider ID and adding a template-provider fallback; fix flash-lite prefix ordering. Treat the fix as a boundary condition and retain the source wording beside your local result.

Keep the full v2026.3.28 notes with your test result. Read the official release index for neighboring versions and the model provider reference for configuration limits.