feat(providers): add GLM-5.3 and GLM-5.3-Flash to Z.ai

[waleedlatif1]

Summary

• Add glm-5.3 (1M context, 131k max output, $1.40/$4.40 per 1M) with reasoning-effort levels low/high/max — reasoning is always on for this model, so no thinking on/off toggle
• Add glm-5.3-flash (1M context, 131k max output, $0.15/$0.50 per 1M), marked speedOptimized
• Move recommended from glm-5.2 to glm-5.3
• Update the Z.ai model-list assertion and regenerate the agent-block streamed-thinking table

Flash pricing uses Z.ai's list rate, not the current 50% promo ($0.075/$0.25) that expires 2026-09-09 — otherwise the entry would silently under-bill once the promo ends.

glm-5.3-flash intentionally omits reasoningEffort: Z.ai's docs recommend reasoning_effort: max for it but never publish the allowed values, so the effort selector stays hidden rather than offering guessed levels.

Type of Change

• New feature

Testing

Tested manually. bun run lint, bun run type-check, bun run check:audits (33/33 including agent-stream-docs:check and check:api-validation:strict), and providers/models.test.ts (38/38) all pass.

Checklist

• Code follows project style guidelines
• Self-reviewed my changes
• Tests added/updated and passing
• No new warnings introduced
• I confirm that I have read and agree to the terms outlined in the Contributor License Agreement (CLA)

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
docs	Building	Preview	Aug 26, 2026 5:08pm

[greptile-apps]

Greptile Summary

Adds GLM-5.3 and GLM-5.3-Flash to the Z.ai provider catalog and updates associated tests and generated documentation.

• Defines pricing, context windows, output limits, release metadata, and model-specific capabilities.
• Moves the Z.ai recommendation from GLM-5.2 to GLM-5.3.
• Updates model registration tests and streamed-thinking documentation.

Confidence Score: 5/5

The PR appears safe to merge with no concrete blocking or non-blocking issues identified.

The new models follow existing provider registration and routing paths, their capability metadata is consistent with request sanitization and generated documentation, and unsupported Pi execution paths remain intentionally guarded.

Important Files Changed

Filename	Overview
apps/sim/providers/models.ts	Adds both Z.ai model definitions with routing, pricing, capability, recommendation, and release metadata; no actionable defect was established.
apps/sim/providers/models.test.ts	Extends the expected Z.ai model list and context-window assertions for both new models.
apps/docs/content/docs/en/workflows/blocks/agent.mdx	Updates the generated streamed-thinking table to include reasoning-capable GLM-5.3 while consistently excluding GLM-5.3-Flash.

_{Reviews (1): Last reviewed commit: "feat(providers): add GLM-5.3 and GLM-5.3..." | Re-trigger Greptile}