Zhipu AI model product
GLM 5.3
3 is a text-only large language model from Zhipu AI for coding, agentic work, and cybersecurity tasks.
Updated Oct 23, 2026. Default version: GLM-5.3
LLMBoard Score84.3GLM-5.3
Coverage80%17 benchmark families
Context window1MTokens
Official input price$1.2Zhipu AI API
GLM 5.3 Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
GLM 5.3 Benchmark Results
Benchmark scores for GLM-5.3.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkPostTrainBench | Score39.80% | Rank02 | Participants9 | Percentile87.50% | EvidenceC | Evaluated |
| BenchmarkTerminal-Bench 3.0 | Score28.30% | Rank02 | Participants4 | Percentile66.67% | EvidenceC | Evaluated |
| BenchmarkFrontierSWE | Score78.10% | Rank03 | Participants17 | Percentile87.50% | EvidenceC | Evaluated |
| BenchmarkGDPval-AA | Score1,769.00 points | Rank03 | Participants17 | Percentile87.50% | EvidenceC | Evaluated |
| BenchmarkExploitBench | Score54.40% | Rank04 | Participants8 | Percentile57.14% | EvidenceC | Evaluated |
| BenchmarkHumanity's Last Exam | Score62.50% | Rank05 | Participants109 | Percentile96.30% | EvidenceC | Evaluated |
| BenchmarkExploitGym | Score15.00% | Rank06 | Participants8 | Percentile28.57% | EvidenceC | Evaluated |
| BenchmarkNL2Repo | Score58.00% | Rank06 | Participants25 | Percentile79.17% | EvidenceC | Evaluated |
| BenchmarkSWE-Marathon | Score42.50% | Rank06 | Participants10 | Percentile44.44% | EvidenceC | Evaluated |
| BenchmarkLM Arena Agent Tool Hallucination | Score0.66% | Rank06 | Participants44 | Percentile88.37% | EvidenceA | Evaluated |
| BenchmarkAA SciCode Subtasks | Score59.03% | Rank06 | Participants89 | Percentile94.32% | EvidenceB | EvaluatedN/A |
| BenchmarkCyberGym | Score84.50% | Rank07 | Participants23 | Percentile72.73% | EvidenceC | Evaluated |
| BenchmarkTerminal-Bench 2.1 | Score88.20% | Rank07 | Participants48 | Percentile87.23% | EvidenceC | Evaluated |
| BenchmarkToolathlon | Score73.00% | Rank08 | Participants42 | Percentile82.93% | EvidenceC | Evaluated |
| BenchmarkAutomationBench | Score48.20% | Rank09 | Participants27 | Percentile69.23% | EvidenceC | Evaluated |
| BenchmarkLM Arena Agent Praise Complaint | Score4.91% | Rank09 | Participants44 | Percentile81.40% | EvidenceA | Evaluated |
| BenchmarkTerminal-Bench 4.0 | Score41.80% | Rank12 | Participants29 | Percentile60.71% | EvidenceB | Evaluated |
| BenchmarkLM Arena Agent Task Outcome Explicit | Score3.37% | Rank15 | Participants44 | Percentile67.44% | EvidenceA | Evaluated |
| BenchmarkLM Arena Webdev | Score1,621.19 rating | Rank15 | Participants107 | Percentile86.79% | EvidenceA | Evaluated |
| BenchmarkAA CritPt | Score19.14% | Rank16 | Participants200 | Percentile92.46% | EvidenceB | EvaluatedN/A |
| BenchmarkAgents' Last Exam | Score28.50% | Rank17 | Participants24 | Percentile30.43% | EvidenceC | Evaluated |
| BenchmarkProgram Bench | Score19.00% | Rank17 | Participants18 | Percentile5.88% | EvidenceC | Evaluated |
| BenchmarkLM Arena Agent Leaderboard | Score1.41% | Rank19 | Participants44 | Percentile58.14% | EvidenceA | Evaluated |
| BenchmarkLiveBench reasoning (2026-06-25) | Score85.80 score | Rank19 | Participants43 | Percentile57.14% | EvidenceB | EvaluatedN/A |
| BenchmarkLM Arena Text | Score1,473.69 rating | Rank22 | Participants218 | Percentile90.32% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Style Control | Score1,480.72 rating | Rank22 | Participants218 | Percentile90.32% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Steerability | Score-0.29% | Rank23 | Participants44 | Percentile48.84% | EvidenceA | Evaluated |
| BenchmarkAA HLE Text No Tools | Score42.26% | Rank23 | Participants200 | Percentile88.94% | EvidenceB | EvaluatedN/A |
| BenchmarkAA GPQA Diamond | Score91.72% | Rank24 | Participants199 | Percentile88.38% | EvidenceB | EvaluatedN/A |
| BenchmarkLiveBench math (2026-06-25) | Score87.90 score | Rank25 | Participants43 | Percentile42.86% | EvidenceB | EvaluatedN/A |
GLM 5.3 Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenatext factuality | Categoryfrench | Rank04 | Rating / score1,518.89 | Votes571 | ObservationsN/A | Result date |
| Arenatext style control | Categoryfrench | Rank04 | Rating / score1,517.33 | Votes819 | ObservationsN/A | Result date |
| Arenaagent tool hallucination | Categoryoverall | Rank06 | Rating / score0.01 | VotesN/A | Observations8.9M | Result date |
| Arenatext | Categoryfrench | Rank08 | Rating / score1,502.57 | Votes819 | ObservationsN/A | Result date |
| Arenaagent praise complaint | Categoryoverall | Rank09 | Rating / score0.05 | VotesN/A | Observations27.6K | Result date |
| Arenatext style control | Categorygerman | Rank11 | Rating / score1,500.70 | Votes382 | ObservationsN/A | Result date |
| Arenatext | Categorychinese | Rank12 | Rating / score1,532.12 | Votes1,845 | ObservationsN/A | Result date |
| Arenatext style control | Categorychinese | Rank12 | Rating / score1,537.92 | Votes1,845 | ObservationsN/A | Result date |
| Arenatext style control | Categorykorean | Rank12 | Rating / score1,463.69 | Votes404 | ObservationsN/A | Result date |
| Arenatext | Categorykorean | Rank13 | Rating / score1,459.88 | Votes404 | ObservationsN/A | Result date |
| Arenawebdev | Categorywebdev-react | Rank13 | Rating / score1,641.06 | Votes6,344 | ObservationsN/A | Result date |
| Arenatext | Categoryexpert | Rank14 | Rating / score1,513.80 | Votes2,358 | ObservationsN/A | Result date |
| Arenatext | Categorygerman | Rank14 | Rating / score1,488.02 | Votes382 | ObservationsN/A | Result date |
| Arenatext | Categoryhard prompts english | Rank14 | Rating / score1,495.85 | Votes6,109 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry medicine and healthcare | Rank14 | Rating / score1,501.81 | Votes1,438 | ObservationsN/A | Result date |
| Arenatext style control | Categoryhard prompts english | Rank14 | Rating / score1,510.44 | Votes6,109 | ObservationsN/A | Result date |
| Arenaagent task outcome explicit | Categoryoverall | Rank15 | Rating / score0.03 | VotesN/A | Observations53.2K | Result date |
| Arenatext | Categoryenglish | Rank15 | Rating / score1,482.31 | Votes9,669 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry life and physical and social science | Rank15 | Rating / score1,498.33 | Votes4,055 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry mathematical | Rank15 | Rating / score1,496.76 | Votes1,272 | ObservationsN/A | Result date |
| Arenatext | Categoryinstruction following | Rank15 | Rating / score1,479.40 | Votes8,022 | ObservationsN/A | Result date |
| Arenatext factuality | Categorychinese | Rank15 | Rating / score1,526.76 | Votes1,488 | ObservationsN/A | Result date |
| Arenatext style control | Categoryenglish | Rank15 | Rating / score1,490.39 | Votes9,669 | ObservationsN/A | Result date |
| Arenatext style control | Categoryexpert | Rank15 | Rating / score1,517.59 | Votes2,358 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry life and physical and social science | Rank15 | Rating / score1,507.67 | Votes4,055 | ObservationsN/A | Result date |
| Arenawebdev | Categoryoverall | Rank15 | Rating / score1,621.19 | Votes9,828 | ObservationsN/A | Result date |
| Arenawebdev | Categorywebdev | Rank15 | Rating / score1,621.19 | Votes9,828 | ObservationsN/A | Result date |
| Arenatext | Categoryhard prompts | Rank16 | Rating / score1,489.66 | Votes15,149 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry medicine and healthcare | Rank17 | Rating / score1,480.76 | Votes1,783 | ObservationsN/A | Result date |
| Arenatext | Categorylonger query | Rank17 | Rating / score1,482.72 | Votes10,923 | ObservationsN/A | Result date |
GLM 5.3 Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $1.2 input, $4.18 output per 1M
- Official provider
- Zhipu AI
- Lowest third-party
- From $0.2857 input, $3.43 output per 1M via Kenari
- Tracked offerings
- 86
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderVolcengine Ark Coding Plan | Provider model IDglm-5.3 | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context1M | Updated |
| ProviderZhipu AI Coding Plan | Provider model IDglm-5.3 | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context1M | Updated |
| ProviderZ.AI Coding Plan | Provider model IDglm-5.3 | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context1M | Updated |
| ProviderSCNet Token Plan | Provider model IDGLM-5.3 | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context1M | Updated |
| ProviderNaN | Provider model IDglm5.3 | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context1M | Updated |
| ProviderAlibaba Token Plan | Provider model IDglm-5.3 | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context1M | Updated |
| ProviderNvidia | Provider model IDz-ai/glm-5.3 | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context1M | Updated |
| ProviderAlibaba Token Plan (China) | Provider model IDglm-5.3 | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context1M | Updated |
| ProviderUmans AI Coding Plan | Provider model IDumans-glm-5.3 | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context1M | Updated |
| ProviderKenari | Provider model IDglm-5-3 | Regionglobal | Input / 1M$0.2857 | Output / 1M$3.43 | Context1M | Updated |
| ProviderCloudFerro Sherlock | Provider model IDzai-org/GLM-5.3 | Regionglobal | Input / 1M$0.34 | Output / 1M$4.48 | Context262.1K | Updated |
| ProviderInceptron | Provider model IDzai-org/GLM-5.3 | Regionglobal | Input / 1M$0.45 | Output / 1M$1.96 | Context1M | Updated |
| ProviderDevPass (LLM Gateway) | Provider model IDglm-5.3 | Regionglobal | Input / 1M$0.50 | Output / 1M$2 | Context1M | Updated |
| ProviderCortecs | Provider model IDglm-5.3 | Regionglobal | Input / 1M$0.511 | Output / 1M$2.05 | Context1M | Updated |
| ProviderNanoGPT | Provider model IDz-ai/glm-5.3 | Regionglobal | Input / 1M$0.60 | Output / 1M$1.1 | Context1M | Updated |
| ProviderOpenRouter | Provider model IDz-ai/glm-5.3 | Regionglobal | Input / 1M$0.70 | Output / 1M$2.2 | Context1M | Updated |
| ProviderMerge Gateway | Provider model IDzai/glm-5.3 | Regionglobal | Input / 1M$0.70 | Output / 1M$2.2 | Context1M | Updated |
| ProviderVultr | Provider model IDglm-5.3 | Regionglobal | Input / 1M$0.75 | Output / 1M$3 | Context1M | Updated |
| ProviderIteraCompute | Provider model IDz-ai/glm-5.3 | Regionglobal | Input / 1M$0.82 | Output / 1M$2.88 | Context1M | Updated |
| ProviderDeep Infra | Provider model IDzai-org/GLM-5.3 | Regionglobal | Input / 1M$0.90 | Output / 1M$4 | Context1M | Updated |
| Providerengy | Provider model IDglm-5.3 | Regionglobal | Input / 1M$0.98 | Output / 1M$3.08 | Context327.7K | Updated |
| Providerai& | Provider model IDzai-org/glm-5.3 | Regionglobal | Input / 1M$1 | Output / 1M$4 | Context1M | Updated |
| ProviderAlibaba (China) | Provider model IDglm-5.3 | Regionglobal | Input / 1M$1.1 | Output / 1M$3.85 | Context1M | Updated |
| ProviderVancine | Provider model IDglm-5.3 | Regionglobal | Input / 1M$1.12 | Output / 1M$3.52 | Context1M | Updated |
| ProviderBoundless | Provider model IDglm-5.3 | Regionglobal | Input / 1M$1.12 | Output / 1M$3.52 | Context1M | Updated |
| ProviderAKI.IO | Provider model IDglm5.3-754b | Regionglobal | Input / 1M$1.12 | Output / 1M$3.93 | Context524.3K | Updated |
| ProviderAIHubMix | Provider model IDglm-5.3 | Regionglobal | Input / 1M$1.13 | Output / 1M$3.94 | Context1M | Updated |
| ProviderMelious | Provider model IDglm-5.3 | Regionglobal | Input / 1M$1.16 | Output / 1M$3.48 | Context1M | Updated |
| ProviderZhipu AI | Provider model IDglm-5.3 | Regionglobal | Input / 1M$1.2 | Output / 1M$4.18 | Context1M | Updated |
| ProviderCrossModel | Provider model IDz-ai/glm-5.3 | Regionglobal | Input / 1M$1.2 | Output / 1M$4.4 | Context1M | Updated |
GLM 5.3 Runtime Performance
Provider-specific output speed and catalog latency for GLM-5.3. Runtime does not affect the capability score.
No runtime data
No provider-specific speed or latency record is linked to the default version yet.
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
GLM 5.3 Specifications
Technical details for the model's default version.
- Version
- GLM-5.3
- Released
- Aug 14, 2026
- Knowledge cutoff
- Unknown
- Parameters
- 753B
- Context window
- 1M
- Max output
- 1M
- Inputs
- text
- Outputs
- text
- Open weights
- Yes
- License
- GLM-5.3 License
GLM 5.3 Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionGLM-5.3 | Released | LLMBoard84.35 | Parameters753B | Context1M | Max output1M | Open weightsYes | LicenseGLM-5.3 License |
Models similar to GLM 5.3
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
ZA
GLM 5.3 Flash
Zhipu AI
ZA
GLM 5.2
Zhipu AI
ZA
GLM 5.1
Zhipu AI
ZA
GLM 5
Zhipu AI
ZA
GLM 4.7
Zhipu AI
ZA
GLM 5V Turbo
Zhipu AI
What is GLM 5.3?
Key information about GLM 5.3 and its available data.
2. It supports a 1M-token context window, 128K maximum output, function calling, structured output, context caching, and configurable reasoning effort levels.
Data as of 2026-10-22.
FAQ
Common questions about GLM 5.3.
When was GLM 5.3 released?
GLM 5.3's default version was released on Aug 14, 2026.
How much does GLM 5.3 cost?
GLM 5.3's official API price is $1.2 per million input tokens and $4.18 per million output tokens via Zhipu AI. The lowest tracked third-party offer starts at $0.2857 input and $3.43 output via Kenari.
Who created GLM 5.3?
GLM 5.3 was created by Zhipu AI.
What is the context window for GLM 5.3?
The default version has a 1M token context window.
Is GLM 5.3 open weight?
Yes. The default version is marked as open weight under GLM-5.3 License.
How many API providers offer GLM 5.3?
86 provider offerings are linked to the default version.