Add OpenRouter pricing for z-ai/glm-5.3-flash

[jlui17]

TLDR

Problem this solves:

• openrouter/z-ai/glm-5.3-flash has no pricing row, so spend logs $0 or a fallback estimate
• The direct zai/glm-5.3-flash row exists, but not the OpenRouter one

How it solves it:

• Adds openrouter/z-ai/glm-5.3-flash to model_prices_and_context_window.json and the backup copy
• Prices are Z.ai's list prices, matching the existing zai/glm-5.3-flash row (OpenRouter is running a temporary 50% promo through 2026-09-09; not encoded)
• Context and max output are from https://openrouter.ai/api/v1/models

User Flow

Before: a developer routing GLM 5.3 Flash through OpenRouter sees no cost for those calls

1. They send POST https://litellm-domain/v1/chat/completions with "model": "openrouter/z-ai/glm-5.3-flash"
2. The response comes back fine, but the spend recorded for the request is $0 because the model has no pricing row
3. They open https://litellm-domain/ui/?page=logs and see the request at $0 spend

After: the same request records real spend

1. They send the same POST with "model": "openrouter/z-ai/glm-5.3-flash"
2. The response comes back the same, and spend is computed from $0.15/M input, $0.50/M output, $0.03/M cache read
3. They open https://litellm-domain/ui/?page=logs and see the request with a real dollar figure

Caveats

• OpenRouter currently shows a limited-time 50% discount for this model (through September 9, 2026). This row uses the undiscounted list price, consistent with the zai/glm-5.3-flash row.

Type

🆕 New Feature

Changes

• model_prices_and_context_window.json: add openrouter/z-ai/glm-5.3-flash
• litellm/model_prices_and_context_window_backup.json: same row

[CLAassistant]

All committers have signed the CLA.

[greptile-apps]

Greptile Summary

Adds OpenRouter pricing, token limits, and capability metadata for z-ai/glm-5.3-flash in both model registries

• Uses the existing direct Z.AI model prices
• Keeps the primary and backup registries synchronized
• Needs functional cost-calculation coverage required for new features

Confidence Score: 4/5

The metadata looks consistent, but the repository’s required functional coverage should be added before merge

The pricing and limits follow existing registry conventions, but this new spend-calculation behavior has no test proving the model resolves and calculates costs correctly

Files Needing Attention: model_prices_and_context_window.json, litellm/model_prices_and_context_window_backup.json

Important Files Changed

Filename	Overview
model_prices_and_context_window.json	Adds consistent OpenRouter model metadata but lacks required functional pricing coverage
litellm/model_prices_and_context_window_backup.json	Mirrors the new model metadata from the primary registry

_{Reviews (1): Last reviewed commit: "Add OpenRouter pricing for z-ai/glm-5.3-..." | Re-trigger Greptile}

[greptile-apps]

Missing functional pricing test

This adds pricing behavior without a functional cost-calculation test. Repository guidance requires meaningful tests for new features, so this requirement must be satisfied before merging

Context Used: CLAUDE.md (source)

Note: If this suggestion doesn't match your team's coding style, reply to this and let me know. I'll remember it for next time!

[codspeed]

Merging this PR will not alter performance

✅ 31 untouched benchmarks

---

_{Comparing jlui17:add-openrouter-glm-5.3-pricing (f8dd867) with main (7d5b645)}

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[devin-ai-integration]

Superseded by the rolling registry PR #31884. The openrouter/z-ai/glm-5.3-flash entry already exists on litellm_internal_staging with the live OpenRouter API prices (7.5e-8 input, 2.5e-7 output, 1.5e-8 cache read), which differ from the values proposed here. Thanks for flagging the gap.