fix(inference): migrate GLM-5 model references to GLM-5.1 (#1744)

[ColinM-sys]

Summary

Swap every curated z-ai/glm5 reference to z-ai/glm-5.1 ahead of the May 9, 2026 full-deprecation date. The successor is already served at the same provider prefix, so this is a straight id/label rename.

Fixes #1744.

Files changed

• src/lib/inference-config.ts — CLOUD_MODEL_OPTIONS picker entry
• src/lib/inference-config.test.ts — id-list assertion
• src/lib/inventory-commands.test.ts — sandbox-list fixtures
• test/onboard-selection.test.ts — mocked /v1/models response + selection answers + expected model
• docs/inference/inference-options.md — NVIDIA Endpoints example list
• .agents/skills/nemoclaw-user-configure-inference/SKILL.md
• .agents/skills/nemoclaw-user-configure-inference/references/inference-options.md

Why now

Per the deprecation notice (issue #1744):

• Apr 9, 2026 — Deprecation notice + banner live (gateway redirect active during migration window)
• May 9, 2026 — GLM-5 fully shut down
• Successor: GLM-5.1

Landing this before May 9 ensures the curated picker and docs no longer suggest a model that will return hard errors post-T0.

Migration note for existing users

Users with sandboxes already pinned to z-ai/glm5 will keep working via the gateway redirect through the migration window. A rebuild after this PR lands will pick up the new default automatically.

Test plan

• npm run build:cli — clean
• npx vitest run src/lib/inference-config.test.ts src/lib/inventory-commands.test.ts — 22/22 pass
• No remaining glm5 or GLM-5 (pre-.1) references anywhere in the tree
• No code-path changes — purely a string rename

Signed-off-by: Colin McDonough cmcdonough@50words.com

Summary by CodeRabbit

• Updates
  • Replaced NVIDIA Endpoints curated model GLM-5 with GLM-5.1 across configuration, tests, onboarding/selection flows, documentation, and quickstart examples to keep model identifiers consistent.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: 2033de34-1a1d-4363-9046-507a5ff36f0c

📥 Commits

Reviewing files that changed from the base of the PR and between 334300d and bc932c0.

📒 Files selected for processing (8)

• .agents/skills/nemoclaw-user-configure-inference/references/inference-options.md
• .agents/skills/nemoclaw-user-get-started/SKILL.md
• docs/get-started/quickstart.md
• docs/inference/inference-options.md
• src/lib/inference-config.test.ts
• src/lib/inference-config.ts
• src/lib/inventory-commands.test.ts
• test/onboard-selection.test.ts

✅ Files skipped from review due to trivial changes (5)

• src/lib/inference-config.test.ts
• .agents/skills/nemoclaw-user-configure-inference/references/inference-options.md
• docs/inference/inference-options.md
• src/lib/inventory-commands.test.ts
• test/onboard-selection.test.ts

🚧 Files skipped from review as they are similar to previous changes (1)

• src/lib/inference-config.ts

---

📝 Walkthrough

Walkthrough

This pull request updates the NVIDIA Endpoints curated model identifier from z-ai/glm5 to z-ai/glm-5.1 across config, tests, onboarding, and docs. The code export CLOUD_MODEL_OPTIONS now uses id: "z-ai/glm-5.1" while the visible label remains GLM-5 in code; documentation examples were updated to GLM-5.1.

Changes

Cohort / File(s)	Summary
Core config src/lib/inference-config.ts	Replaced cloud model id z-ai/glm5 with z-ai/glm-5.1 in CLOUD_MODEL_OPTIONS (label in code kept as "GLM-5").
Config & inventory tests src/lib/inference-config.test.ts, src/lib/inventory-commands.test.ts	Updated expected model id strings from z-ai/glm5 to z-ai/glm-5.1.
Onboarding tests test/onboard-selection.test.ts	Mocked /v1/models response and assertions updated to use z-ai/glm-5.1.
Reference docs .agents/skills/nemoclaw-user-configure-inference/references/inference-options.md, docs/inference/inference-options.md	Provider options table updated: curated model entry changed from GLM-5 to GLM-5.1.
User guidance / Quickstart .agents/skills/nemoclaw-user-get-started/SKILL.md, docs/get-started/quickstart.md	Examples and onboarding wizard guidance updated to show GLM-5.1 in model selection lists.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~10 minutes

Poem

🐰 A hop, a tweak, a tiny fix so neat,
IDs aligned and docs now sing complete.
GLM-5.1 trots in, a polite little bow,
Tests and guides agree — migration's done now. 🥕

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately describes the main change: migrating GLM-5 model references to GLM-5.1 across the codebase.
Linked Issues check	✅ Passed	The PR fully satisfies issue #1744 requirements by replacing all GLM-5 references with GLM-5.1 in curated pickers, documentation, and test fixtures.
Out of Scope Changes check	✅ Passed	All changes are scoped to updating GLM-5 to GLM-5.1 across documentation and code; no unrelated modifications are present.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[wscurran]

✨ Thanks for submitting this PR that proposes a fix and enhancement to the inference capabilities of NemoClaw by migrating GLM-5 model references to GLM-5.1, which could help ensure compatibility with the latest models.

---

Possibly related open issues:

• #1744 Model Deprecation – [API Catalog] [glm5] [10 Days Notice]

[copy-pr-bot]

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.