Z.ai GLM 5.3 API Pricing and Speed
Z.ai GLM 5.3 API pricing: $1.40 per 1M tokens for input and $4.40 per 1M tokens for output, from the official API price list of Z.ai. Official rate checked (source); OpenRouter is a separate channel below. Our measured latency and output speed are shown by route and country below. Published OpenRouter: Last checked · Source (data) · Methodology
OpenRouter: Amount worked out from the source page: $0.07 per 1M tokens, in USD · as shown in the United States on 9 Nov 2026 · the page does not say whether tax is included; we do not assume it is excluded · last price change seen on 28 Oct 2026.
Price changed 28 Oct 2026. input $0.079 → $0.0699 (−11.5 %) per 1M tokens on OpenRouter. Confirmed by a later check on 29 Oct 2026. Briefing: what it does to task costs
Quality on your task: not measured by Measured Lab yet. On our data, GLM 5.3 fits 5 of 6 common tasks by published prices, context window and terms, with calculated monthly costs.
What can I use it for?
Six common tasks cost a month at list prices
- Calculated Answer customer messages: $25.20 for 5,000 customer messages a month
- Unknown Pull data from invoices and documents with strict JSON and page images: not confirmed, reads text only, per the OpenRouter listing
- Calculated Summarize documents and meetings: $12.42 for 1,000 documents or meetings a month
- Calculated Translate content: $16.08 for 3,000 pages a month
- Calculated Chat assistant on your website: $268.50 for 10,000 conversations a month
- Calculated Write and review code with tool calls: $77.05 for 500 coding tasks a month
What our data confirms
- Observed Strict JSON output: accepted in our test call on 17 Oct 2026
- Published Calls your tools, in Z.ai documentation, read 17 Oct 2026
- Published Reads text only; OpenRouter listing, 9 Nov 2026
- Published Weights on Hugging Face: zai-org/GLM-5.3, as linked by the OpenRouter listing
Example to try
“Review this pull request and list bugs and missing tests, with the file and line of each.” Run it on your own data before you choose.
What we measured
- Observed Speed on a short test request: first token 1.9 s, 71 tokens/s (median of 114 test requests on 23 dates, short request)
Stated by the provider
- Stated by the provider Z.ai for this model: Terminal-Bench 2.1 88.2, Terminal-Bench 3.0 28.3, DeepSWE v1.1 66.9; quotes and test conditions in Quality signals
Before you choose
- Unknown Training on your data: terms not in our data
- Unknown Where requests are processed: Unknown
Prices USD per 1M tokens, as listed
- $1.40Input per 1M tokens, from Z.ai Published
- $4.40Output per 1M tokens, from Z.ai Published
- $0.26Cached input per 1M tokens, from Z.ai Published
OpenRouter listing $0.0699 / $1.40 per 1M tokens: listing differs from 44 of 45 endpoints read on 9 Nov 2026.
All prices by channel
| Price per 1M tokens | Z.ai direct | Through OpenRouter | Previous prices by channel |
|---|---|---|---|
| Input | $1.40 Published | $0.0699–$2.80 Publishedacross 45 endpoints; listing $0.0699: listing differs from 44 of 45 endpoints | OpenRouter: |
| Output | $4.40 Published | $1.40–$8.80 Publishedacross 45 endpoints; listing $1.40: listing differs from 44 of 45 endpoints | OpenRouter: |
| Cached input · read | $0.26 Published | $0.0689 Published | — |
| Batch · input / output | — | $0.45 / $2.00 Published | — |
Input price from Z.ai, 24 Oct 2026 – 9 Nov 2026
24 Oct 2026$1.40 throughout9 Nov 2026
Input price through OpenRouter, 16 Oct 2026 – 9 Nov 2026
16 Oct 2026$0.079 → $0.0699 (7 changes)9 Nov 2026
Context window1,048,576tokens Published
Max output943,718tokens Published
OpenRouter providers who serves this model through OpenRouter, at what price
OpenRouter sends a call without a pinned provider to one of these endpoints, and each endpoint has its own price. Prices are USD per 1M tokens, input / output.
Pinned min: $0.0699 / $1.40 · Wafer (wafer/us)
Unpinned range: input $0.0699–$2.80, output $1.40–$8.80 across 45 endpoints of 34 providers (listing shows $1.00 / $4.20).
No endpoint in this reading sells at the listing price.
Price moves of 20% or more since the reading of 8 Nov 2026:
- Listing price $0.0699 / $1.40 → $1.00 / $4.20 (input +1331%, output +200%).
- Morph (morph/fp8): $0.74 / $3.00 → $1.245 / $2.542 (input +68%, output −15%).
Observed unpinned cost: 144 short requests sent without a pinned provider on 25 Oct 2026 – 9 Nov 2026 (16 dates), in each routing mode. Charged is the sum of the OpenRouter generation receipts for these requests; the listing price column is the price of the same input and output tokens at the listing price of that date.
| Routing mode | Requests | Dates | Served by | Charged | Listing price for the same tokens | Charged / listing | Lowest-priced endpoint answered |
|---|---|---|---|---|---|---|---|
| Default (no provider field) | 48 | 25 Oct 2026 – 9 Nov 2026 (16 dates) | 36 of 48 Wafer (wafer/us), 8 of 48 Wafer (wafer/fp4), 4 of 48 Inceptron (inceptron/fp4) | $0.0011 | $0.0023 | 0.50× | 36 of 48 |
:floor | 48 | 25 Oct 2026 – 9 Nov 2026 (16 dates) | 42 of 48 Wafer (wafer/us), 6 of 48 Relace (relace) | $0.00089 | $0.0019 | 0.47× | 42 of 48 |
:nitro | 48 | 25 Oct 2026 – 9 Nov 2026 (16 dates) | 17 of 48 BaseTen (baseten/us/fast), 11 of 48 Fireworks (fireworks/fast), 9 of 48 Alibaba (alibaba/prime), 7 of 48 Modal (modal), 2 of 48 Parasail (parasail/fast), 1 of 48 Amazon Bedrock (amazon-bedrock/us-east-1), 1 of 48 InferenceNet (inference-net/us) | $0.0100 | $0.0040 | 2.49× | 0 of 48 |
Routing modes as documented by OpenRouter (OpenRouter provider routing, read 25 Oct 2026): the default request has no provider field; :floor and :nitro are suffixes of the model name. Served by names the endpoint that answered, by the provider name and endpoint tag of the receipt. Each request asks for one short line with an answer cap of 32 tokens. The counts are for these requests on these dates only.
Models with the widest provider price spread on OpenRouter
Request fragment that pins Wafer (wafer/us):
{"provider": {"order": ["wafer/us"], "allow_fallbacks": false}}With fallbacks turned off, OpenRouter does not move the call to another provider. A full endpoint tag pins that endpoint; the provider name alone covers all of its endpoints (OpenRouter provider routing).
All endpoints (45), by input price
| Provider | Endpoint tag | Precision | Input | Output | Context, tokens | Status code |
|---|---|---|---|---|---|---|
| Wafer | wafer/us | fp4 | $0.0699 | $1.40 | 1,048,576 | 0 |
| Relace | relace/fp4 | fp4 | $0.0703 | $4.3843 | 1,048,576 | 0 |
| InferenceNet | inference-net/fp4 | fp4 | $0.08 | $3.60 | 1,048,576 | 0 |
| InferenceNet | inference-net/us | fp4 | $0.08 | $3.60 | 1,048,576 | 0 |
| Reka | reka | Not stated | $0.14 | $2.10 | 1,048,576 | 0 |
| Wafer | wafer/fp4 | fp4 | $0.179 | $1.40 | 1,048,576 | 0 |
| DekaLLM | dekallm | Not stated | $0.18 | $2.95 | 1,048,576 | 0 |
| AkashML | akashml/fp4 | fp4 | $0.19 | $4.40 | 1,048,576 | 0 |
| Makora | makora/fp4 | fp4 | $0.30 | $3.00 | 1,048,576 | 0 |
| PrimeIntellect | primeintellect | Not stated | $0.50 | $3.08 | 1,048,576 | 0 |
| Decart | decart/fp4 | fp4 | $0.56 | $1.78 | 1,048,576 | 0 |
| DeepInfra | deepinfra/fp4 | fp4 | $0.5625 | $2.50 | 1,048,576 | -2 |
| Novita | novita/fp8 | fp8 | $0.63 | $1.98 | 1,048,576 | 0 |
| Cloudflare | cloudflare | Not stated | $0.70 | $2.20 | 1,048,576 | 0 |
| GMICloud | gmicloud/fp8 | fp8 | $0.70 | $2.20 | 1,048,576 | 0 |
| SiliconFlow | siliconflow/fp8 | fp8 | $0.70 | $2.20 | 1,048,576 | 0 |
| CoreWeave | coreweave/nvfp4 | nvfp4 | $0.75 | $3.00 | 1,048,576 | 0 |
| Sail Research | sail-research/us | fp8 | $0.75 | $3.00 | 1,048,576 | 0 |
| Phala | phala | Not stated | $0.84 | $2.64 | 1,048,576 | 0 |
| DigitalOcean | digitalocean | Not stated | $0.91 | $2.86 | 1,048,576 | 0 |
| Fireworks | fireworks/spot | Not stated | $0.98 | $3.08 | 1,048,576 | 0 |
| Inceptron | inceptron/fp4 | fp4 | $1.00 | $4.90 | 1,048,576 | 0 |
| Sail Research | sail-research/fp8 | fp8 | $1.05 | $4.20 | 1,048,576 | 0 |
| Alibaba | alibaba | Not stated | $1.19 | $3.74 | 1,000,000 | 0 |
| Morph | morph/fp8 | fp8 | $1.245 | $2.542 | 1,048,576 | 0 |
| Friendli | friendli | Not stated | $1.26 | $3.96 | 1,048,576 | 0 |
| AtlasCloud | atlas-cloud/fp8 | fp8 | $1.40 | $4.40 | 1,048,576 | 0 |
| Baidu | baidu/fp8 | fp8 | $1.40 | $4.40 | 1,048,576 | 0 |
| BaseTen | baseten/fp4 | fp4 | $1.40 | $4.40 | 1,048,576 | 0 |
| BaseTen | baseten/fp4/us | fp4 | $1.40 | $4.40 | 1,048,576 | 0 |
| Crusoe | crusoe/fp4 | fp4 | $1.40 | $4.40 | 1,048,576 | 0 |
| Fireworks | fireworks | Not stated | $1.40 | $4.40 | 1,048,576 | 0 |
| Mistral | mistral/zdr | nvfp4 | $1.40 | $4.40 | 1,048,576 | 0 |
| Modal | modal | Not stated | $1.40 | $4.40 | 1,048,576 | 0 |
| Parasail | parasail/fp8 | fp8 | $1.40 | $4.40 | 1,048,576 | 0 |
| Together | together | Not stated | $1.40 | $4.40 | 1,048,575 | 0 |
| Venice | venice | Not stated | $1.40 | $4.40 | 1,000,000 | 0 |
| Z.AI | z-ai/fp8 | fp8 | $1.40 | $4.40 | 1,048,576 | 0 |
| Mistral | mistral/nvfp4/eu | nvfp4 | $1.54 | $4.84 | 1,048,576 | 0 |
| Amazon Bedrock | amazon-bedrock/us-east-1 | Not stated | $1.68 | $5.28 | 1,048,576 | 0 |
5 more endpoints in the source.
Sorted by input price, then output price. Precision is the quantization field of the endpoints list. Status code is the status field of the endpoints list, as read.
Measured by us our own short test request: about 380 tokens in, a short JSON answer out · medians from answers on two dates or more
United States
| Route | First answer token | Output speed | Cost per request | Hidden reasoning |
|---|---|---|---|---|
| Direct API · Z.aireasoning: low114 answers on 23 dates · 99 % answered | 1.93 srange 0.85–7.37 s · slow 10 %: 3.34 s Observed | 70.7 tokens/srange 26.8–471 · slowest 10 %: 36.4 tokens/s | $0.000822range $0.000744–$0.00101 Calculated | 0for 70 visible |
| OpenRouter → Z.AIreasoning: low115 answers on 23 dates · 100 % answered | 2.19 srange 1.09–32.48 s · slow 10 %: 3.17 s Observed | 69.7 tokens/srange 22.8–610 · slowest 10 %: 40.9 tokens/s | $0.000374range $0.000307–$0.000576 Calculated | 0for 69 visible |
United Kingdom
| Route | First answer token | Output speed | Cost per request | Hidden reasoning |
|---|---|---|---|---|
| Direct API · Z.aireasoning: low105 answers on 21 dates · 100 % answered | 1.83 srange 0.76–6.82 s · slow 10 %: 3.28 s Observed | 72.1 tokens/srange 20.7–587 · slowest 10 %: 41.1 tokens/s | $0.000828range $0.00076–$0.00171 Calculated | 0for 71 visible |
| OpenRouter → Z.AIreasoning: low105 answers on 21 dates · 100 % answered | 1.88 srange 0.81–9.85 s · slow 10 %: 3.25 s Observed | 68.1 tokens/srange 13.8–494 · slowest 10 %: 30.8 tokens/s | $0.000365range $0.000312–$0.000639 Calculated | 0for 70 visible |
Germany
| Route | First answer token | Output speed | Cost per request | Hidden reasoning |
|---|---|---|---|---|
| Direct API · Z.aireasoning: low104 answers on 21 dates · 99 % answered | 1.69 srange 0.66–11.20 s · slow 10 %: 3.45 s Observed | 69.1 tokens/srange 29.2–619 · slowest 10 %: 36.8 tokens/s | $0.000825range $0.000763–$0.00159 Calculated | 0for 68 visible |
| OpenRouter → Z.AIreasoning: low105 answers on 21 dates · 100 % answered | 1.79 srange 0.84–20.81 s · slow 10 %: 3.04 s Observed | 75.1 tokens/srange 30.2–546 · slowest 10 %: 44.8 tokens/s | $0.000379range $0.000307–$0.00106 Calculated | 0for 70 visible |
Canada
| Route | First answer token | Output speed | Cost per request | Hidden reasoning |
|---|---|---|---|---|
| Direct API · Z.aireasoning: low105 answers on 21 dates · 100 % answered | 2.13 srange 0.89–6.04 s · slow 10 %: 3.51 s Observed | 75.3 tokens/srange 27.9–504 · slowest 10 %: 46.6 tokens/s | $0.000834range $0.000743–$0.00169 Calculated | 0for 71 visible |
| OpenRouter → Z.AIreasoning: low104 answers on 21 dates · 100 % answered | 2.17 srange 1.13–9.71 s · slow 10 %: 3.59 s Observed | 71.6 tokens/srange 17.4–427 · slowest 10 %: 44.4 tokens/s | $0.000375range $0.000312–$0.00121 Calculated | 0for 70 visible |
Australia
| Route | First answer token | Output speed | Cost per request | Hidden reasoning |
|---|---|---|---|---|
| Direct API · Z.aireasoning: low105 answers on 21 dates · 100 % answered | 2.21 srange 1.16–9.77 s · slow 10 %: 3.32 s Observed | 70.8 tokens/srange 24.4–662 · slowest 10 %: 41.9 tokens/s | $0.000831range $0.00075–$0.001 Calculated | 0for 70 visible |
| OpenRouter → Z.AIreasoning: low105 answers on 21 dates · 100 % answered | 2.60 srange 1.28–12.10 s · slow 10 %: 3.73 s Observed | 75.8 tokens/srange 36.1–482 · slowest 10 %: 53.1 tokens/s | $0.000374range $0.000307–$0.00117 Calculated | 0for 70 visible |
- Hidden reasoning tokens are billed as output, at the output price, even though they are not part of the answer.
- Through OpenRouter, the request goes to the named provider; the price and speed can differ from the direct API.
How the test works and what each number means: methodology. The figures are also in our open data.
Quality signals public test results, each with its source and date
| Signal | Result | Source |
|---|---|---|
| Instruction followingIFEval, 50 prompts · reasoning: low | 92 % Observedof 50 prompts with every instruction followed | Our run of a fixed IFEval subset, checked by code92 % on 22 Oct 2026 · 94 % on 17 Oct 2026 · instruction level 94 % |
Figures the provider publishes (6)
| Benchmark | Provider's figure | Where the provider says it |
|---|---|---|
| Terminal-Bench 2.1 | 88.2 Stated by the provider | Results table in zai-org/GLM-5.3 · Hugging Face (README, Benchmark)read 17 Oct 2026 |
| Terminal-Bench 3.0 | 28.3 Stated by the provider | Results table in zai-org/GLM-5.3 · Hugging Face (README, Benchmark)read 17 Oct 2026 |
| DeepSWE v1.1 | 66.9 Stated by the provider | Results table in zai-org/GLM-5.3 · Hugging Face (README, Benchmark)read 17 Oct 2026 |
| Humanity's Last Exam (HLE w/ Tools) | 62.5 Stated by the provider | Results table in zai-org/GLM-5.3 · Hugging Face (README, Benchmark)read 17 Oct 2026 |
| CyberGym | 84.5% Stated by the provider | On CyberGym, which starts from white-box source code and tests whether the model can identify and validate vulnerabilities by triggering faults, GLM-5.3 scores 84.5%, up from GLM-5.2’s 77.2%GLM-5.3 - Overview - Z.AI DEVELOPER DOCUMENTread 17 Oct 2026 |
| SWE-Marathon v1.1 | 42.5 Stated by the provider | Results table in zai-org/GLM-5.3 · Hugging Face (README, Benchmark)read 17 Oct 2026 |
- Your own task: quality on it is not measured here. These are public test sets; a short trial on a few of your own examples is the check that fits your work.
- Other public leaderboards, without figures copied here: human preference votes on open-ended answers: Arena (LMArena) text leaderboard; long tasks in coding, data analysis and research: LiveBench.
What each signal covers and how our run is scored: methodology.
How we got this number
- What we read: the published price on openrouter.ai/api/v1/models
- Checked: ; tracked since 16 Oct 2026
- Shown in: the United States
- Checks so far: 25 checks since 16 Oct 2026
- What changed: the price last changed on ; every change is listed under History
- Page updated:
Other measured facts
| Question | Answer | Checked | Status |
|---|---|---|---|
| How much does GLM-5.3 cost per 1M cache-read tokens (cache hits) directly from Z.ai (official API price)? | $0.26 per 1M tokens | Published | |
| How much does GLM-5.3 cost per 1M input tokens directly from Z.ai (official API price)? | $1.40 per 1M tokens | Published | |
| How much does GLM-5.3 cost per 1M output tokens directly from Z.ai (official API price)? | $4.40 per 1M tokens | Published | |
| How much does Z.ai GLM 5.3 cost per 1M cache-read tokens (cache hits) on OpenRouter? | $0.07 per 1M tokens | Published | |
| How much does Z.ai GLM 5.3 cost per 1M output tokens on OpenRouter? | $1.40 per 1M tokens | Published | |
| What is the context window of Z.ai GLM 5.3 on OpenRouter, in tokens? | 1,048,576 tokens | Published | |
| What is the maximum output length of Z.ai GLM 5.3 on OpenRouter, in tokens? | 943,718 tokens | Published |
History
- — $0.08 per 1M tokens → $0.07 per 1M tokens Confirmed by the check on 29 Oct 2026. What this means for you: If you buy or renew this plan now, you pay the new price; the old price is shown for comparison.
- — $4.80 per 1M tokens → $1.40 per 1M tokens (How much does Z.ai GLM 5.3 cost per 1M output tokens on OpenRouter?) Confirmed by the check on 20 Oct 2026. What this means for you: If you buy or renew this plan now, you pay the new price; the old price is shown for comparison.
- — First check: $0.08 per 1M tokens
- Short-lived or unconfirmed moves are not shown as changes.
The source record lists $0.00000007 per token ($0.07 per 1M tokens), fetched on 9 Nov 2026.
The source record, as published
"prompt":"0.0000000699"
Briefings on this model each price change recalculated as a task cost
Record history — 164 records for 8 questions
- Last record
- Shown in
- the United States
| Time | Value | Status |
|---|---|---|
| 9 Nov 2026 | $0.07 per 1M tokens | Published |
| 8 Nov 2026 | $0.07 per 1M tokens | Published |
| 7 Nov 2026 | $0.07 per 1M tokens | Published |
| 6 Nov 2026 | $0.07 per 1M tokens | Published |
| 5 Nov 2026 | $0.07 per 1M tokens | Published |
| 4 Nov 2026 | $0.08 per 1M tokens | Published |
| 3 Nov 2026 | $0.08 per 1M tokens | Published |
| Time | Value | Status |
|---|---|---|
| 9 Nov 2026 | $0.26 per 1M tokens | Published |
| 8 Nov 2026 | $0.26 per 1M tokens | Published |
| 7 Nov 2026 | $0.26 per 1M tokens | Published |
| 6 Nov 2026 | $0.26 per 1M tokens | Published |
| 5 Nov 2026 | $0.26 per 1M tokens | Published |
| 4 Nov 2026 | $0.26 per 1M tokens | Published |
| 3 Nov 2026 | $0.26 per 1M tokens | Published |
| 2 Nov 2026 | $0.26 per 1M tokens | Published |
| Time | Value | Status |
|---|---|---|
| 9 Nov 2026 | $1.40 per 1M tokens | Published |
| 8 Nov 2026 | $1.40 per 1M tokens | Published |
| 7 Nov 2026 | $1.40 per 1M tokens | Published |
| 6 Nov 2026 | $1.40 per 1M tokens | Published |
| 5 Nov 2026 | $1.40 per 1M tokens | Published |
| 4 Nov 2026 | $1.40 per 1M tokens | Published |
| 3 Nov 2026 | $1.40 per 1M tokens | Published |
| 2 Nov 2026 | $1.40 per 1M tokens | Published |
| Time | Value | Status |
|---|---|---|
| 9 Nov 2026 | $4.40 per 1M tokens | Published |
| 8 Nov 2026 | $4.40 per 1M tokens | Published |
| 7 Nov 2026 | $4.40 per 1M tokens | Published |
| 6 Nov 2026 | $4.40 per 1M tokens | Published |
| 5 Nov 2026 | $4.40 per 1M tokens | Published |
| 4 Nov 2026 | $4.40 per 1M tokens | Published |
| 3 Nov 2026 | $4.40 per 1M tokens | Published |
| Time | Value | Status |
|---|---|---|
| 9 Nov 2026 | $0.07 per 1M tokens | Published |
| 8 Nov 2026 | $0.07 per 1M tokens | Published |
| 7 Nov 2026 | $0.07 per 1M tokens | Published |
| 6 Nov 2026 | $0.07 per 1M tokens | Published |
| 5 Nov 2026 | $0.07 per 1M tokens | Published |
| 4 Nov 2026 | $0.08 per 1M tokens | Published |
| 3 Nov 2026 | $0.08 per 1M tokens | Published |
| 2 Nov 2026 | $0.08 per 1M tokens | Published |
| Time | Value | Status |
|---|---|---|
| 9 Nov 2026 | $1.40 per 1M tokens | Published |
| 8 Nov 2026 | $1.40 per 1M tokens | Published |
| 7 Nov 2026 | $1.40 per 1M tokens | Published |
| 6 Nov 2026 | $1.40 per 1M tokens | Published |
| 5 Nov 2026 | $1.40 per 1M tokens | Published |
| 4 Nov 2026 | $3.60 per 1M tokens | Published |
| 3 Nov 2026 | $3.60 per 1M tokens | Published |
| Time | Value | Status |
|---|---|---|
| 9 Nov 2026 | 1,048,576 tokens | Published |
| 8 Nov 2026 | 1,048,576 tokens | Published |
| 7 Nov 2026 | 1,048,576 tokens | Published |
| 6 Nov 2026 | 1,048,576 tokens | Published |
| 5 Nov 2026 | 1,048,576 tokens | Published |
| 4 Nov 2026 | 1,048,576 tokens | Published |
| 3 Nov 2026 | 1,048,576 tokens | Published |
| Time | Value | Status |
|---|---|---|
| 9 Nov 2026 | 943,718 tokens | Published |
| 8 Nov 2026 | 943,718 tokens | Published |
| 7 Nov 2026 | 943,718 tokens | Published |
| 6 Nov 2026 | 943,718 tokens | Published |
| 5 Nov 2026 | 943,718 tokens | Published |
| 4 Nov 2026 | 943,718 tokens | Published |
| 3 Nov 2026 | 943,718 tokens | Published |
| 2 Nov 2026 | 943,718 tokens | Published |
Showing the latest 60 of 164 records; the full history is in data.json and data.csv.