GLM 5.2 API Prices Rise 50%: Input Now $0.84, Output $2.64 per 1M Tokens

What Changed

Z.ai has increased GLM 5.2 API prices by exactly +50% across both input and output. Input tokens move from $0.56 to $0.84 per 1M tokens; output tokens jump from $1.76 to $2.64 per 1M tokens. No new capabilities or model version were announced alongside the change.

Does This Matter for Your Workload?

For low-volume or experimental use, the absolute dollar impact stays modest — a workload burning 10M output tokens monthly goes from roughly $17.60 to $26.40. That's noticeable but not catastrophic.

The math gets less comfortable at scale. High-throughput applications — summarization pipelines, document processing, customer-facing chat — will see the full 50% hit on output costs, which typically dominate total spend. Output pricing at $2.64/1M now puts GLM 5.2 closer to mid-tier Western models rather than the budget-friendly positioning it previously held.

What to Do Next

If GLM 5.2 is a core part of your stack, it's worth benchmarking your actual input/output ratio. Output-heavy workloads feel this increase most. Input-heavy workloads (e.g., long-context retrieval or classification) still get hit, but proportionally less.

Before re-signing or scaling up, check comparable models on the site. Several alternatives sit at or below the old GLM 5.2 price points — track current rates and run a side-by-side on GLM 5.2 — live specs & price history.

Bottom Line

A 50% price increase with no accompanying capability bump is a straightforward cost-of-doing-business change. It doesn't make GLM 5.2 unusable, but it removes the pricing edge that made it attractive for cost-sensitive deployments. Re-evaluate if output volume is high.

Ezra, Scout AI Team