// AI Platform Drift
Z.ai GLM 5.3 batch: output limit cut to 131K tokens, pricing up 3-9%
OpenRouter model catalogue · published 2026-09-22 16:28 UTC
WHAT CHANGEDThe Z.ai GLM 5.3 batch model reduced max_completion_tokens from 943,718 to 131,072 tokens (86% reduction). Prompt pricing increased from $0.0000007 to $0.00000072 per token (2.9% increase). Completion pricing increased from $0.0000022 to $0.0000024 per token (9.1% increase). Context length remains 1,048,576 tokens.
WHY IT MATTERSAny batch job generating outputs longer than 131K tokens will now fail or truncate. If you were relying on the previous 943K output capacity for long-form generation, summarization of large documents, or multi-document synthesis, those workflows will break. The pricing increase affects cost forecasts for high-volume batch processing.
WHAT TO DOGrep your codebase for 'glm-5.3:batch' and check max_tokens or completion length assumptions. If any job expects outputs over 131K tokens, either chunk the work into multiple calls or switch models. Recalculate batch processing costs using the new rates and update budgets if this model handles significant volume.
Model-reported confidence 95% (not measured accuracy)
Primary source ↗
written by anthropic/claude-sonnet-4.5
Show the reasoning — raw diff, triage verdict, confidence
RAW DIFF
id: z-ai/glm-5.3:batch name: Z.ai: GLM 5.3 (batch) -pricing.prompt: 0.0000007 -pricing.completion: 0.0000022 +pricing.prompt: 0.00000072 +pricing.completion: 0.0000024 context_length: 1048576 -top_provider.max_completion_tokens: 943718 +top_provider.max_completion_tokens: 131072
HOW IT WAS JUDGED
Triage signals below are model-reported assertions, preserved verbatim; they are not independent arithmetic verification.
| triage verdict | MATERIAL at 0.95, needed 0.5 |
|---|---|
| triage reasoning | Pricing increased and max output tokens cut dramatically, affecting cost and feasibility of workloads. |
| signals | pricing.prompt increased 2.9% · pricing.completion increased 9.1% · max_completion_tokens decreased 86% (943718 → 131072) · breaking change to output capacity |
| suppression | — |
| triage model | anthropic/claude-haiku-4.5 ($0.00000) |
| writer model | anthropic/claude-sonnet-4.5 ($0.00000) |