z-ai/glm-5.3

模型发布 OpenRouter 模型 原文语言:中文

OpenRouter 于 2026-08-19 收录该模型(收录时间不等于模型正式发布日期);上下文:1,048,576 tokens;输入价:$0.0840/百万 tokens;输出价:$3.6000/百万 tokens;定位(来源描述):GLM-5.3 is a large-scale reasoning model from Z.ai, built for complex software engineering and…

全文

Z.ai: GLM 5.3

z-ai/glm-5.3

Model weights

Compare

GLM-5.3 is a large-scale reasoning model from Z.ai, built for complex software engineering and long-horizon agent tasks. It supports text input and output with a 1M-token context window, and improves on GLM-5.2 in coding and in the balance between performance and token efficiency.

Reasoning is always on and cannot be disabled. Reasoning efforts low, high, and max are supported; max is the default.

Modalities

In / out price

$0.0699 / $1.40per 1M

Context

1.0M

Released

Aug 18, 2026

Compare

Providers

Different companies host the same model. OpenRouter routes your request to one of them based on the routing mode you pick: Balanced (price + speed), Nitro (fastest), or Floor (cheapest).

Latency / throughput

$0.6200$1.9800$0.11000.55s107 tok/s

100.00%

$0.0699$1.4000$0.06890.26s99 tok/s

100.00%

$0.1790$1.4000$0.17800.29s103 tok/s

100.00%

$0.6000$2.0000$0.10000.66s99 tok/s

100.00%

$0.9100$2.8600$0.16900.52s101 tok/s

100.00%

$0.7500$3.0000$0.19000.49s103 tok/s

98.20%

$0.0840$3.6000$0.07900.35s92 tok/s

99.97%

$0.0840$3.6000$0.07900.35s91 tok/s

100.00%

38% off

$0.90$0.5625$4.00$2.5000$0.20$0.12500.61s86 tok/s

99.56%

$1.4000$4.4000$0.26000.27s163 tok/s

99.78%

$1.0000$4.0000$0.16000.32s92 tok/s

100.00%

$1.4000$4.4000$0.26000.50s118 tok/s

100.00%

$1.4000$4.4000$0.26000.43s110 tok/s

100.00%

$1.0000$4.2000$0.13500.70s108 tok/s

99.89%

$0.9800$3.0800$0.18200.56s87 tok/s

99.41%

$1.4000$4.4000$0.14000.72s102 tok/s

100.00%

$1.4000$4.4000$0.14000.72s102 tok/s

99.96%

Fast

$2.1000$6.6000$0.26000.75s144 tok/s

97.46%

$1.4000$4.4000$0.14001.25s109 tok/s

99.95%

$1.2390$3.8120$0.41800.64s72 tok/s

100.00%

$0.1900$4.4000$0.19002.18s97 tok/s

99.92%

Fast

$2.1000$6.6000$0.39001.09s167 tok/s

99.78%

55% off

$1.40$0.6300$4.40$1.9800$0.26$0.11701.73s58 tok/s

100.00%

$0.0703$4.3843$0.07031.54s59 tok/s

100.00%

20% off

$1.75$1.4000$5.50$4.4000$0.325$0.26000.56s115 tok/s

100.00%

10% off

$1.40$1.2600$4.40$3.9600$0.26$0.23400.43s80 tok/s

100.00%

$1.5400$4.8400$0.15401.33s88 tok/s

99.93%

$1.4000$4.4000$0.26001.03s80 tok/s

99.89%

50% off

$1.40$0.7000$4.40$2.2000$0.26$0.13001.62s55 tok/s

99.93%

$0.7000$2.2000$0.13003.93s64 tok/s

100.00%

$1.4000$4.4000$0.26001.71s58 tok/s

100.00%

50% off

$1.40$0.7000$4.40$2.2000$0.26$0.13002.65s50 tok/s

100.00%

$1.6800$5.2800$0.31200.91s95 tok/s

99.65%

Fast

$2.1000$6.6000$0.21000.74s126 tok/s

100.00%

50% off

$2.80$1.4000$8.80$4.4000$0.14$0.07001.20s44 tok/s

100.00%

$1.4000$4.4000$0.26001.65s91 tok/s

95.46%

$1.4000$4.4000$0.26001.29s62 tok/s

100.00%

Prime

$2.8000$8.8000$0.56000.79s107 tok/s

100.00%

$0.7500$3.0000$0.12003.83s20 tok/s

99.93%

$1.0500$4.2000$0.16808.00s16 tok/s

99.82%

$0.9500$3.5000$0.19000.55s94 tok/s

99.63%

$1.1900$3.7400$0.23800.99s71 tok/s

99.50%

Fast

$2.1000$6.6000$0.21000.61s114 tok/s

100.00%

$1.4000$4.4000$0.28000.84s73 tok/s

100.00%

$1.4000$4.4000$0.26001.44s49 tok/s

93.45%

$1.4000$4.4000$0.26006.15s32 tok/s

100.00%

$1.4000$4.4000$0.2600----

98.91%

Not used in Standard routing:Why these endpoints are not used

Flex38% off

$0.72$0.4500$3.20$2.0000$0.16$0.10001.97s73 tok/s

100.00%

ProviderInput / Output /M

Input $0.62 / Output $1.98

107 tok/s100.00% uptime

Input $0.0699 / Output $1.40

99 tok/s100.00% uptime

Input $0.179 / Output $1.40

103 tok/s100.00% uptime

Input $0.60 / Output $2.00

99 tok/s100.00% uptime

Input $0.91 / Output $2.86

101 tok/s100.00% uptime

Input $0.75 / Output $3.00

103 tok/s98.20% uptime

Input $0.084 / Output $3.60

92 tok/s99.97% uptime

Input $0.084 / Output $3.60

91 tok/s100.00% uptime

Input $0.5625 / Output $2.50

38% off

86 tok/s99.56% uptime

Input $1.40 / Output $4.40

163 tok/s99.78% uptime

Input $1.00 / Output $4.00

92 tok/s100.00% uptime

Input $1.40 / Output $4.40

118 tok/s100.00% uptime

Input $1.40 / Output $4.40

110 tok/s100.00% uptime

Input $1.00 / Output $4.20

108 tok/s99.89% uptime

Input $0.98 / Output $3.08

87 tok/s99.41% uptime

Input $1.40 / Output $4.40

102 tok/s100.00% uptime

Input $1.40 / Output $4.40

102 tok/s99.96% uptime

Input $2.10 / Output $6.60

Fast

144 tok/s97.46% uptime

Input $1.40 / Output $4.40

109 tok/s99.95% uptime

Input $1.239 / Output $3.812

72 tok/s100.00% uptime

Input $0.19 / Output $4.40

97 tok/s99.92% uptime

Input $2.10 / Output $6.60

Fast

167 tok/s99.78% uptime

Input $0.63 / Output $1.98

55% off

58 tok/s100.00% uptime

Input $0.0703 / Output $4.3843

59 tok/s100.00% uptime

Input $1.40 / Output $4.40

20% off

115 tok/s100.00% uptime

Input $1.26 / Output $3.96

10% off

80 tok/s100.00% uptime

Input $1.54 / Output $4.84

88 tok/s99.93% uptime

Input $1.40 / Output $4.40

80 tok/s99.89% uptime

Input $0.70 / Output $2.20

50% off

55 tok/s99.93% uptime

Input $0.70 / Output $2.20

64 tok/s100.00% uptime

Input $1.40 / Output $4.40

58 tok/s100.00% uptime

Input $0.70 / Output $2.20

50% off

50 tok/s100.00% uptime

Input $1.68 / Output $5.28

95 tok/s99.65% uptime

Input $2.10 / Output $6.60

Fast

126 tok/s100.00% uptime

Input $1.40 / Output $4.40

50% off

44 tok/s100.00% uptime

Input $1.40 / Output $4.40

91 tok/s95.46% uptime

Input $1.40 / Output $4.40

62 tok/s100.00% uptime

Input $2.80 / Output $8.80

Prime

107 tok/s100.00% uptime

Input $0.75 / Output $3.00

20 tok/s99.93% uptime

Input $1.05 / Output $4.20

16 tok/s99.82% uptime

Input $0.95 / Output $3.50

94 tok/s99.63% uptime

Input $1.19 / Output $3.74

71 tok/s99.50% uptime

Input $2.10 / Output $6.60

Fast

114 tok/s100.00% uptime

Input $1.40 / Output $4.40

73 tok/s100.00% uptime

Input $1.40 / Output $4.40

49 tok/s93.45% uptime

Input $1.40 / Output $4.40

32 tok/s100.00% uptime

Input $1.40 / Output $4.40

-- tok/s98.91% uptime

Not used in Standard routing:Why these endpoints are not used

Input $0.45 / Output $2.00

Flex

+11 more label

73 tok/s100.00% uptime

Pricing

The average price customers actually pay for this model, next to the prices providers post. Caching and discounts mean the price actually paid is often well below the listed one.

Weighted average

Input

$0.1816/M tokens

Output

$3.028/M tokens

GLM 5.3 — Price history

Price history

No effective price data for this selection

$0.0000$0.000.0%0.0%

$0.0690$1.4091.3%24.0%

$0.0703$4.3888.6%1.4%

$0.0794$3.6092.8%5.3%

$0.0794$3.6092.6%7.0%

$0.1353$1.8192.4%12.5%

$0.1456$2.0090.9%7.0%

$0.1742$2.2092.2%0.4%

$0.1760$2.2192.0%0.9%

$0.1775$2.5390.0%1.2%

$0.1781$1.4090.6%5.5%

$0.1900$4.4070.5%0.4%

$0.1920$4.4095.9%0.6%

$0.1964$4.2093.3%2.8%

$0.1993$3.0087.4%0.4%

$0.2020$1.9883.5%1.4%

$0.2036$2.2087.1%0.2%

$0.2225$3.5095.7%0.8%

$0.2517$3.0089.0%3.0%

$0.2661$2.8686.9%1.5%

$0.2699$4.4089.7%0.8%

$0.2755$3.0888.3%1.6%

$0.2894$4.2086.2%0.0%

$0.2950$4.0083.9%0.5%

$0.2965$3.7493.9%1.4%

$0.3002$4.4082.7%0.0%

$0.3308$4.4084.9%0.4%

$0.3351$4.4093.4%8.7%

$0.3530$4.8485.6%0.2%

$0.3543$4.4093.4%0.0%

$0.3736$4.4090.0%2.6%

Fast

$0.3843$6.6090.8%0.3%

$0.3877$4.4088.8%0.1%

$0.4043$4.4087.3%0.6%

Fast

$0.4061$6.6089.6%0.4%

$0.4109$4.4086.8%1.5%

Fast

$0.4227$6.6091.2%0.6%

$0.4248$4.4185.6%0.1%

$0.4393$4.4084.3%0.1%

$0.4410$3.2486.1%0.1%

$0.4615$4.4082.3%0.0%

$0.4641$4.4082.1%0.1%

Fast

$0.5584$6.6090.2%2.5%

$0.5690$4.0567.4%0.2%

$0.5742$5.2880.8%0.2%

Prime

$0.7461$8.8091.7%0.3%

$0.7788$4.4054.5%0.0%

Not used in Standard routing:Why these endpoints are not used

Flex

$0.1757$2.0078.4%0.0%

ProviderInput / Output /M

Token share

Effective in $0.0000 / Effective out $0.00

Token share

0.0%

Effective in $0.0690 / Effective out $1.40

Token share

24.0%

Effective in $0.0703 / Effective out $4.38

Token share

1.4%

Effective in $0.0794 / Effective out $3.60

Token share

5.3%

Effective in $0.0794 / Effective out $3.60

Token share

7.0%

Effective in $0.1353 / Effective out $1.81

Token share

12.5%

Effective in $0.1456 / Effective out $2.00

Token share

7.0%

Effective in $0.1742 / Effective out $2.20

Token share

0.4%

Effective in $0.1760 / Effective out $2.21

Token share

0.9%

Effective in $0.1775 / Effective out $2.53

Token share

1.2%

Effective in $0.1781 / Effective out $1.40

Token share

5.5%

Effective in $0.1900 / Effective out $4.40

Token share

0.4%

Effective in $0.1920 / Effective out $4.40

Token share

0.6%

Effective in $0.1964 / Effective out $4.20

Token share

2.8%

Effective in $0.1993 / Effective out $3.00

Token share

0.4%

Effective in $0.2020 / Effective out $1.98

Token share

1.4%

Effective in $0.2036 / Effective out $2.20

Token share

0.2%

Effective in $0.2225 / Effective out $3.50

Token share

0.8%

Effective in $0.2517 / Effective out $3.00

Token share

3.0%

Effective in $0.2661 / Effective out $2.86

Token share

1.5%

Effective in $0.2699 / Effective out $4.40

Token share

0.8%

Effective in $0.2755 / Effective out $3.08

Token share

1.6%

Effective in $0.2894 / Effective out $4.20

Token share

0.0%

Effective in $0.2950 / Effective out $4.00

Token share

0.5%

Effective in $0.2965 / Effective out $3.74

Token share

1.4%

Effective in $0.3002 / Effective out $4.40

Token share

0.0%

Effective in $0.3308 / Effective out $4.40

Token share

0.4%

Effective in $0.3351 / Effective out $4.40

Token share

8.7%

Effective in $0.3530 / Effective out $4.84

Token share

0.2%

Effective in $0.3543 / Effective out $4.40

Token share

0.0%

Effective in $0.3736 / Effective out $4.40

Token share

2.6%

Effective in $0.3843 / Effective out $6.60

FastToken share

0.3%

Effective in $0.3877 / Effective out $4.40

Token share

0.1%

Effective in $0.4043 / Effective out $4.40

Token share

0.6%

Effective in $0.4061 / Effective out $6.60

FastToken share

0.4%

Effective in $0.4109 / Effective out $4.40

Token share

1.5%

Effective in $0.4227 / Effective out $6.60

FastToken share

0.6%

Effective in $0.4248 / Effective out $4.41

Token share

0.1%

Effective in $0.4393 / Effective out $4.40

Token share

0.1%

Effective in $0.4410 / Effective out $3.24

Token share

0.1%

Effective in $0.4615 / Effective out $4.40

Token share

0.0%

Effective in $0.4641 / Effective out $4.40

Token share

0.1%

Effective in $0.5584 / Effective out $6.60

FastToken share

2.5%

Effective in $0.5690 / Effective out $4.05

Token share

0.2%

Effective in $0.5742 / Effective out $5.28

Token share

0.2%

Effective in $0.7461 / Effective out $8.80

PrimeToken share

0.3%

Effective in $0.7788 / Effective out $4.40

Token share

0.0%

Not used in Standard routing:Why these endpoints are not used

Effective in $0.1757 / Effective out $2.00

FlexToken share

0.0%

Performance

Throughput is how fast the model writes (tokens per second — higher is better). Latency is total round-trip time (lower is better). TTFT is time-to-first-token — how long before you see anything appear (lower is better).

Throughput

167tok/s

P50, best across providers

Latency

0.26s

P50, best provider

Reasoning effortLatency / throughput

AutoExacto benchmarks

GPQA DiamondTAU-BenchBaseten (fp8) (US)84.7%83.7%Wafer88.4%78.7%Wafer (US)87.6%79.1%Doubleword88.6%78.0%Z.ai88.0%77.4%DeepInfra87.0%78.1%Baseten (fp4) (US)84.0%76.6%Baseten (fp4)84.1%76.2%Baseten (fp8)84.6%70.4%

+39 more providers

Uptime

Uptime is the percentage of the past 3 days that at least one provider was responding to requests. Availability is the percentage of time that inference was successfully served. OpenRouter continuously monitors and uses the next-best provider when one returns an error.

Uptime (3d)The model was reachable. Request routed to a provider.

99.99%

Availability (3d)The model returned inference from any provider. Errors and empty responses count against it.

99.93%

Availability over the last 3 days

Last 72 hours

Availability 99.93%

3 Days Ago2 Days AgoYesterdayNow

Availability over the last 24 hours

OpenRouter Availability

99.95%

Without Routing

97.09%

When an error occurs in an upstream provider, we can recover by routing to another healthy provider, if your request filters allow it. You can access per-provider uptime data programmatically through the Endpoints API. Learn more about our load balancing and provider routing options.

Benchmarks

Scores on standardized evaluations and this model’s own outputs on shared benchmark prompts. Higher percentages are better — and each rank shows where this model lands among the models evaluated on that source.

Benchmark score summary for Z.ai: GLM 5.3 (Artificial Analysis, Design Arena and Vals AI)

SourceBenchmarkScore

Artificial AnalysisGLM-5.3 (Max) Intelligence Index31.7

Artificial AnalysisGLM-5.3 (Max) Coding Index74.8

Artificial AnalysisGLM-5.3 (Max) Agentic Index53.1

Artificial AnalysisGLM-5.3 (Max) GPQA Diamond91.7%

Artificial AnalysisGLM-5.3 (Max) HLE42.3%

Artificial AnalysisGLM-5.3 (Max) AA-LCR79.7%

Artificial AnalysisGLM-5.3 (Max) τ-Bench Banking50.3%

Artificial AnalysisGLM-5.3 (Max) GDPval-AA57.7%

Artificial AnalysisGLM-5.3 (Max) CritPt19.1%

Artificial AnalysisGLM-5.3 (Max) SciCode59.0%

Artificial AnalysisGLM-5.3 (Max) Terminal-Bench 4.041.9%

Artificial AnalysisGLM-5.3 (Max) Terminal-Bench 2.183.9%

Artificial AnalysisGLM-5.3 (Max) AA-Omniscience Accuracy32.3%

Artificial AnalysisGLM-5.3 (Max) AA-Omniscience Non-Hallucination Rate61.1%

Artificial AnalysisGLM-5.3 (Low) Intelligence Index23.1

Artificial AnalysisGLM-5.3 (Low) HLE36.6%

Artificial AnalysisGLM-5.3 (Low) AA-LCR72.7%

Artificial AnalysisGLM-5.3 (Low) GDPval-AA41.6%

Artificial AnalysisGLM-5.3 (Low) CritPt14.6%

Artificial AnalysisGLM-5.3 (Low) SciCode42.0%

Artificial AnalysisGLM-5.3 (Low) Terminal-Bench 4.034.8%

Artificial AnalysisGLM-5.3 (Low) AA-Omniscience Accuracy31.9%

Artificial AnalysisGLM-5.3 (Low) AA-Omniscience Non-Hallucination Rate36.8%

Design ArenaGLM-5.3 Models Arena 3D Elo1294

Design ArenaGLM-5.3 Models Arena Asciiart Elo1184

Design ArenaGLM-5.3 Models Arena Code Categories Elo1283

Design ArenaGLM-5.3 Models Arena Data Visualization Elo1251

Design ArenaGLM-5.3 Models Arena Game Development Elo1288

Design ArenaGLM-5.3 Models Arena SVG Elo1245

Design ArenaGLM-5.3 Models Arena UI Component Elo1283

Design ArenaGLM-5.3 Models Arena Website Elo1279

Design ArenaGLM-5.3 Agents Arena Htmlslides Elo1189

Design ArenaGLM-5.3 Agents Arena Mobile Apps Elo1184

Design ArenaGLM-5.3 Agents Arena Python-Pptxslides Elo1217

Design ArenaGLM-5.3 Agents Arena Webapps Elo1270

Vals AIGLM 5.3 Access to Justice Benchmark73.7%

Vals AIGLM 5.3 Code Migration v1.155.1%

Vals AIGLM 5.3 CyberBench v1.172.1%

Vals AIGLM 5.3 EMB v1.169.1%

Vals AIGLM 5.3 Finance Agent v2.165.0%

Vals AIGLM 5.3 Vals Finance Index61.4%

Vals AIGLM 5.3 Financial Coaching Bench74.1%

Vals AIGLM 5.3 GPQA Diamond88.1%

Vals AIGLM 5.3 HLAB-Vals-1.114.6%

Vals AIGLM 5.3 Immigration66.7%

Vals AIGLM 5.3 IOI68.4%

Vals AIGLM 5.3 LiveCodeBench80.5%

Vals AIGLM 5.3 LegalBench84.8%

Vals AIGLM 5.3 Vals Legal Index74.6%

Vals AIGLM 5.3 Legal Research Bench49.0%

Vals AIGLM 5.3 MedCode42.9%

Vals AIGLM 5.3 MedScribe88.8%

Vals AIGLM 5.3 MMLU Pro86.8%

Vals AIGLM 5.3 MysteryMechanism23.0%

Vals AIGLM 5.3 ProgramBench1.5%

Vals AIGLM 5.3 ProofBench v1.149.0%

Vals AIGLM 5.3 Public Benefits Bench v1.168.5%

Vals AIGLM 5.3 Vals RSI Index22.9%

Vals AIGLM 5.3 SkillsBench47.5%

Vals AIGLM 5.3 SRE Bench9.2%

Vals AIGLM 5.3 SWE-bench95.4%

Vals AIGLM 5.3 Tax Agent Bench43.3%

Vals AIGLM 5.3 TaxEval v272.4%

Vals AIGLM 5.3 Terminal-Bench 4.038.9%

Vals AIGLM 5.3 Terminal-Bench-Science5.7%

Vals AIGLM 5.3 Vals Index59.8%

Vals AIGLM 5.3 Vibe Code Bench 1-100 v1.129.8%

Vals AIGLM 5.3 Vibe Code Bench v1.178.1%

Apps

Public apps that send the most traffic to this model. Good signal for what real production workloads look like — and a hint at which use cases this model is best suited for.

1.

Claude Code

Claude Code is Anthropic's agentic coding tool that reads your entire codebase, plans and executes changes across files, runs tests, and iterates on failures, all from natural language prompts.

1.85Ttokens

2.

Hermes Agent

Hermes Agent is an open-source, self-improving AI agent by Nous Research that runs persistently with memory across sessions, and builds reusable skills from experience. It comes with 40+ built-in tools, including web search, browser automation, and vision, plus scheduled automations and subagents.

905Btokens

3.

pi

There are many coding agents, but this one is yours.

697Btokens

4.

Strix

Open-source AI hackers that find and validate vulnerabilities in your code, APIs, and infrastructure, and open fix PRs.

581Btokens

5.

omp

new

473Btokens

Rest of OpenRouter

211B

Top 5 Apps

83.6B

Activity

Token volume and request traffic to this model over time.

API

Drop-in code to call this model. OpenRouter's API is OpenAI-compatible — most SDKs work by just swapping the base URL. The only thing that changes between models is the model slug below.

Z.ai: GLM versions

Available through EU in-region routing

Z.ai: GLM 5.3 is available through EU in-region routing. Requests sent to eu.openrouter.ai are decrypted in the EU and routed only to shared provider endpoints in Europe (the EU/EEA, the UK, and Switzerland).

Providers in EuropeInceptron, MistralPrices in Europe

Inceptron: $1/M tokens input, $4.20/M tokens output

Mistral: $1.54/M tokens input, $4.84/M tokens output

Base URLhttps://eu.openrouter.ai/api/v1PlanRequires Business or Enterprise. Self-serve on Business: no contract, no monthly minimum. OpenRouter Business for EU in-region routing

Fails closed: if no shared provider endpoint in Europe is available, the request errors instead of leaving the region. Requests served by your own BYOK key go to your deployment’s region, which you control. In-region routing docs

Documentation

Guides, cookbooks, blog posts, and tools for building with Z.ai: GLM 5.3 on OpenRouter.

Self-Hosted LLM vs API: When Does Self-Hosting Pay OffBlog post

How to Benchmark LLMsBlog post

Frequently asked questions

GLM-5.3 is a large-scale reasoning model from Z.ai, built for complex software engineering and long-horizon agent tasks. It supports text input and output with a 1M-token context window, and improves on GLM-5.2 in coding and in the balance between performance and token efficiency. Reasoning is always on and cannot be disabled. Reasoning efforts low, high, and max are supported; max is the default.

The GLM 5.3 documentation on OpenRouter includes the Self-Hosted LLM vs API: When Does Self-Hosting Pay Off blog post and the How to Benchmark LLMs blog post. The quickstart covers API keys and your first request, and the API reference lists every parameter.

GLM 5.3 costs $0.0699/M input tokens and $1.40/M output tokens, with separate rates for Cache Read at $0.0689/M tokens.

GLM 5.3 has a 1,048,576 token context window. It supports up to 1,048,576 completion tokens.

Yes. GLM 5.3 accepts tools for function calling. It also supports structured outputs via a JSON schema in response_format.

GLM 5.3 is served by 36 providers on OpenRouter: Wafer, Relace, inference.net, AkashML, DeepInfra, Tenstorrent, Decart, NovitaAI and 28 more. Requests are routed to the best available provider, with automatic failover to the others, and you can pin or exclude providers with provider routing.

GLM 5.3 was released on August 18, 2026.

正文以来源语言显示。

参考来源

← 返回每日简报