z-ai
Z.ai: GLM 5.3
z-ai/glm-5.3
Our take
AI-assisted analysis based on the sources below. Suitability is an editorial assessment; test it on your own workload.
z-ai/glm-5.3 is a text-only reasoning model positioned for complex software engineering and long-horizon agent work. Its model card says it shares GLM-5.2's base, with gains coming from post-training, and describes an open-weights release with a 1M-token context window. Reasoning is mandatory and defaults to maximum effort, although high and low settings are available. Vendor-reported benchmarks emphasize coding, terminal, cyber and tool-use tasks, while OpenRouter lists additional catalog evaluations. These results should be treated as source-reported evidence, not independently verified performance.[1][2]
Strengths
Limitations
- Only text input and output are listed; image, audio and video support are not documented.[1]
- Reasoning cannot be disabled in the catalog; only the effort level can be changed.[1][2]
- The pack contains catalog and vendor material only; independent verification, safety-behavior evidence, hardware needs and custom-license terms are absent, so workload validation is necessary.[1][2]
Where to use it
- Editorial suggestion: coding assistants and software-engineering agents for large codebases, after validating correctness on real repositories.[1][2]
- Editorial suggestion: long-horizon tool-use or workflow-automation prototypes that need structured outputs and controllable reasoning effort.[1][2]
- Editorial suggestion: controlled defensive vulnerability-discovery research, given the card's reported cyber benchmarks; real-world safety and reliability remain unproven.[2]
Sources
- [1] OpenRouter catalog · Z.ai: GLM 5.3Sources checked: Nov 13, 2026
- [2] Model documentation · Z.ai: GLM 5.3Sources checked: Nov 13, 2026
About this model
Description below is the original text supplied by OpenRouter, shown in the source language.
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...
Specifications
- Model ID
- z-ai/glm-5.3
- Canonical ID
- z-ai/glm-5.3-20260816
- Creator
- z-ai
- Added to OpenRouter
- 18.08.2026
- Context length
- 1,048,576
- Modality
- text->text
- Input modalities
- text
- Output modalities
- text
- Token pricing
- USD per 1M tokens: in $0.084 / out $3.6
- Knowledge cutoff
- —
- Expiration date
- —
Tasks and capabilities
A grade appears only where the source supplies a matching measurement for that task, the model is active and the cohort holds at least 10 comparable models. The capability column lists facts the source declares — supported formats, reasoning and tool parameters — which are not measured quality.
| Task | Grade | Source-declared capability |
|---|---|---|
| General quality | BB — next 30% of the cohort | reasoning: mandatory · default max · efforts: max, high, low · parameters: include_reasoning, reasoning, reasoning_effort |
| Coding | AA — next 20% of the cohort | — |
| Agents | SS — top 10% of the cohort | parameters: structured_outputs, tool_choice, tools |
| Writing | No measurement | output: text |
| Reasoning | No measurement | reasoning: mandatory · default max · efforts: max, high, low · parameters: include_reasoning, reasoning, reasoning_effort |
| Document analysis | No measurement | — |
| Vision | No measurement | — |
| Image generation | No measurement | — |
| Audio | No measurement | — |
| Video | No measurement | — |
| Embeddings | No measurement | — |
Evaluations
| Source | Benchmark | Value | Grade | Data retrieved |
|---|---|---|---|---|
| Artificial Analysis | Intelligence index | 31.7 | BB — next 30% of the cohort | 13.11.2026 |
| Artificial Analysis | Coding index | 74.8 | AA — next 20% of the cohort | 13.11.2026 |
| Artificial Analysis | Agentic index | 53.1 | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | models · 3dwin rate 56% · rank 19 | 1,299Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | models · asciiartwin rate 47.2% · rank 23 | 1,210Elo | BB — next 30% of the cohort | 13.11.2026 |
| Design Arena | models · codecategorieswin rate 52.3% · rank 21 | 1,285Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | models · datavizwin rate 50.6% · rank 34 | 1,253Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | models · gamedevwin rate 50.6% · rank 21 | 1,290Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | models · svgwin rate 51.4% · rank 17 | 1,248Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | models · uicomponentwin rate 52.8% · rank 26 | 1,284Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | models · websitewin rate 52.6% · rank 23 | 1,280Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | agents · htmlslideswin rate 39.1% · rank 2 | 1,189Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | agents · mobileappswin rate 52.2% · rank 23 | 1,186Elo | BB — next 30% of the cohort | 13.11.2026 |
| Design Arena | agents · python-pptxslideswin rate 50.4% · rank 10 | 1,206Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | agents · webappswin rate 54.2% · rank 10 | 1,270Elo | AA — next 20% of the cohort | 13.11.2026 |
How grades work
Each letter grade compares a model only against other models scored by the same source and metric in the current snapshot. S means the top 10%, A the next 20%, B the next 30%, C the rest. Models without a published measurement are shown as not rated — a supported format alone never implies measured quality.
- S — S — top 10% of the cohort
- A — A — next 20% of the cohort
- B — B — next 30% of the cohort
- C — C — below the B band
Providers
Optional live provider statistics from OpenRouter. Missing values are unknown, not zero.
| Provider | Context length | Price in /1M | Price out /1M | Latency p50 (30m, s) | Throughput p50 (30m, tok/s) | Uptime (30m) |
|---|---|---|---|---|---|---|
| Waferfp4 | 1,048,576 | $0.070 | $1.4 | — | — | 100% |
| Relacefp4 | 1,048,576 | $0.070 | $4.38 | — | — | 100% |
| InferenceNetfp4 | 1,048,576 | $0.084 | $3.6 | — | — | 100% |
| InferenceNetfp4 | 1,048,576 | $0.084 | $3.6 | — | — | 100% |
| Reka | 1,048,576 | $0.150 | $3 | — | — | 100% |
| Waferfp4 | 1,048,576 | $0.179 | $1.4 | — | — | 100% |
| AkashMLfp4 | 1,048,576 | $0.190 | $4.4 | — | — | 100% |
| DeepInfrafp4 | 1,048,576 | $0.563 | $2.5 | — | — | 100% |
| Baidufp8 | 1,048,576 | $0.570 | $1.79 | — | — | 100% |
| Tenstorrentfp8 | 1,044,480 | $0.600 | $2.5 | — | — | 100% |
| Novitafp8 | 1,048,576 | $0.630 | $1.98 | — | — | 100% |
| Morphfp8 | 1,048,576 | $0.643 | $1.98 | — | — | 100% |
| Decartfp4 | 1,048,576 | $0.690 | $2.15 | — | — | 100% |
| SiliconFlowfp8 | 1,048,576 | $0.700 | $2.2 | — | — | 100% |
| GMICloudfp8 | 1,048,576 | $0.700 | $2.2 | — | — | 100% |
| Cloudflare | 1,048,576 | $0.700 | $2.2 | — | — | 100% |
| CoreWeavenvfp4 | 1,048,576 | $0.750 | $3 | — | — | 100% |
| Sail Researchfp8 | 1,048,576 | $0.750 | $3 | — | — | 100% |
| DigitalOcean | 1,048,576 | $0.910 | $2.86 | — | — | 100% |
| Coherenvfp4 | 1,048,576 | $0.950 | $3.5 | — | — | 100% |
| Fireworks | 1,048,576 | $0.980 | $3.08 | — | — | 100% |
| Makorafp4 | 1,048,576 | $1 | $4 | — | — | 100% |
| Inceptronfp4 | 1,048,576 | $1 | $4.2 | — | — | 100% |
| Sail Researchfp8 | 1,048,576 | $1.05 | $4.2 | — | — | 100% |
| Alibaba | 1,000,000 | $1.19 | $3.74 | — | — | 100% |
| Friendli | 1,048,576 | $1.26 | $3.96 | — | — | 100% |
| SambaNova | 1,048,576 | $1.4 | $4.4 | — | — | — |
| Mistralnvfp4 | 1,048,576 | $1.4 | $4.4 | — | — | 100% |
| BaseTenfp4 | 1,048,576 | $1.4 | $4.4 | — | — | 100% |
| Nebiusfp4 | 1,048,576 | $1.4 | $4.4 | — | — | 100% |
| Crusoefp4 | 1,048,576 | $1.4 | $4.4 | — | — | 93.33% |
| Venice | 1,000,000 | $1.4 | $4.4 | — | — | — |
| Together | 1,048,575 | $1.4 | $4.4 | — | — | 100% |
| Parasailfp8 | 1,048,576 | $1.4 | $4.4 | — | — | 100% |
| Modal | 1,048,576 | $1.4 | $4.4 | — | — | 100% |
| BaseTenfp4 | 1,048,576 | $1.4 | $4.4 | — | — | 100% |
| Fireworks | 1,048,576 | $1.4 | $4.4 | — | — | 100% |
| AtlasCloudfp8 | 1,048,576 | $1.4 | $4.4 | — | — | — |
| Z.AIfp8 | 1,048,576 | $1.4 | $4.4 | — | — | 100% |
| Mistralnvfp4 | 1,048,576 | $1.54 | $4.84 | — | — | 100% |
| Amazon Bedrock | 1,048,576 | $1.68 | $5.28 | — | — | 100% |
| Parasailfp8 | 1,048,576 | $2.1 | $6.6 | — | — | 90% |
| Fireworks | 1,048,576 | $2.1 | $6.6 | — | — | 93.33% |
| BaseTenfp8 | 1,048,576 | $2.1 | $6.6 | — | — | 100% |
| BaseTenfp8 | 1,048,576 | $2.1 | $6.6 | — | — | 100% |
| Alibaba | 1,000,000 | $2.8 | $8.8 | — | — | 100% |
Provider data retrieved: 13.11.2026
Related articles
No published articles identify this model yet.