GLM 5.3 API Prices Rise 67%: Input Hits $1.40, Output $4.40 per 1M Tokens

What Changed

Z.ai has raised GLM 5.3 API prices by +67% effective immediately. Input tokens move from $0.84 → $1.40 per 1M tokens; output tokens jump from $2.64 → $4.40 per 1M tokens.

See the full pricing timeline on the GLM 5.3 — live specs & price history page.

Does This Matter for Your Workload?

That depends heavily on your input/output ratio.

• Output-heavy workloads (long-form generation, summarization, agentic chains) take the hardest hit. At $4.40/1M output tokens, a pipeline burning 10M output tokens/month now costs $44 more per month just from this single price change — before any input costs.
• Input-heavy, output-light workloads (classification, embeddings-adjacent tasks, short answers) feel the same proportional increase but in smaller absolute dollar terms.
• Low-volume users — hobby projects or prototypes under a few million tokens/month — will likely absorb the difference without needing to act.

What to Do Next

If GLM 5.3 sits in a cost-sensitive production stack, now is a good time to benchmark alternatives. Several models in a similar capability tier are available on AI Tool Scout at lower per-token rates. Filter by price on our models index and run a side-by-side cost estimate against your actual token split.

If GLM 5.3 is uniquely delivering quality you can't replicate elsewhere, the new pricing may still be justifiable — but it's worth verifying rather than assuming.

No explanation for the increase has been published by Z.ai at the time of writing.

Ezra, Scout AI Team