GLM-5 Turbo API Pricing
Off the rate card: GLM-5 Turbo was Zhipu's latency-optimized GLM-5 variant, and Z.ai no longer lists or prices it. Its row was on the International card on August 24, 2026 and gone by August 29 — the same week GLM-5.3-Flash was added — with no retirement notice published and no entry left in the Z.ai docs index. The figures below are frozen at the last published values: $1.20/M input, $4/M output, $0.24/M on a cache hit. We hold no Z.ai key, so whether the endpoint still answers is not something this page can claim; what it can show is that the vendor stopped selling it. For a current model at this tier see GLM-5.2, or GLM-5.3-Flash at a ninth of the price.
Run the numbers.
Calculator pre-loaded with the last rates Z.ai published for GLM-5 Turbo. They are a historical record, not a quote: the model is off the card, so use this to read old invoices or to size the gap against a model you can still buy.
Real-world presets.
Repo implementation
Pull request review
Knowledge base answer
Support assistant
Paste text. See tokens. See cost.
Counts use a chars-per-token calibration measured on the vendor's own published tokenizer (zai-org/GLM-5, 2026-06-10). English prose is typically within a few percent; code and non-Latin scripts tokenize heavier. For billing-exact counts use the vendor's count-tokens API.
| Model | Input /M | Output /M | Effective blended | Context | Best for |
|---|---|---|---|---|---|
| GLM-5 Turbo Current | $1.20 cache $0.24 | $4.00 | $0.70 agentic 92/8 | 200K | Latency-optimized GLM-5 |
| GLM-5.1 | $1.40 cache $0.26 | $4.40 | $0.78 pricier | 200K | Flagship agentic coding |
| GLM-5 | $1.00 cache $0.20 | $3.20 | $0.572 cheaper | 200K | Cheaper GLM-5 family base |
| GLM-4.7 | $0.60 cache $0.11 | $2.20 | $0.358 cheaper | 200K | Mid-tier GLM agents |
| GLM-4.7 FlashX | $0.07 cache $0.01 | $0.40 | $0.0511 cheaper | 200K | Ultra-cheap GLM traffic |
| GLM-4.7 Flash | Free cache Free | Free | Free cheaper | 200K | Free registered-user tier |
| DeepSeek V4 Pro | $1.32 cache $0.044 | $3.96 | $0.569 cheaper | 1M | DeepSeek frontier discount tier |
| Qwen3 Coder Plus | $1.00 | $5.00 | $1.32 pricier | 1M | Qwen code flagship |
Frequently asked.
Practical pricing questions, separated from calculator assumptions and regional tiers.
Q · 01 What is GLM-5 Turbo priced at? +
$1.20/M input, $4/M output and $0.24/M on a cache hit, and those are the figures frozen on this page. The row was still on the card on August 24, 2026 and gone by August 29.Q · 02 How is the effective price calculated? +
$0.7/M.Q · 03 Is prompt caching priced separately? +
$0.24/M against $1.2/M fresh, a fifth of the rate. Z.ai still applies that structure to the models it does sell — see the GLM lineup — and still shows free cached input storage across the range.Q · 04 Are regional prices different? +
Q · 05 Is there a batch discount? +
Q · 06 How accurate is the tokenizer estimate? +
zhipu-tokenizer-estimate chars-per-token estimate for English text. It is useful for rough planning, but actual billing comes from the vendor API usage fields and can differ for Chinese, code, or mixed-language prompts.