MODEL PROOF
GLM 5.3
FRONTIER
Nothing is both better and cheaper under this workload.
$1.29 / 1M tokens Balanced · 3 tokens in per 1 out
Serving precision differs between offers.
- Independent LMArena score
- 1,475.1 ± 6.4 · 10,960 votes
- Context window
- 1.31M
- Maximum output
- 131.07K
- Input modalities
- text
- Output modalities
- text
- Published input price
- $0.840 / 1M tokens
- Published output price
- $2.64 / 1M tokens
- Pricing kind
- fixed
Inspect complete billing conditions and endpoint terms below.
Every price dimension · USD per million tokens
| Input | $0.840/M |
|---|---|
| Output | $2.64/M |
| Cached input | $0.156/M |
| Cache write | Unknown/M |
| Cache write 1h | Unknown/M |
| Reasoning | Unknown/M |
Who sells it
The same weights, different shops. Cheapest is not like-for-like when serving precision differs.
Headline cheapest is a lower precision. Like-for-like at the best declared precision is $0.561/M from Baidu.
35 provider offers · rates, limits and conditions
Rates are USD per million tokens. Endpoint terms can differ even when the quoted price and precision match. A listed parameter is a provider declaration, not a task-success test.
| Seller | Input | Output | Precision | Uptime (1d) |
|---|---|---|---|---|
| Baidubaidu/fp8 Endpoint terms
| $0.561 | $1.76 | fp8 | 99.94% |
| DeepInfradeepinfra/fp4 Endpoint terms
| $0.563 | $2.50 | fp4 | 95.56% |
| Rekareka/fp8 Endpoint terms
| $0.760 | $2.57 | fp8 | 99.00% |
| Sail Researchsail-research/fp8 Endpoint terms
| $0.770 | $4.00 | fp8 | 99.32% |
| Novitanovita/fp8 Endpoint terms
| $0.783 | $2.46 | fp8 | 99.86% |
| Inceptroninceptron/fp4 Endpoint terms
| $0.786 | $2.65 | fp4 | 99.68% |
| Io Netio-net/fp8 Endpoint terms
| $0.793 | $2.68 | fp8 | 99.95% |
| Phalaphala Endpoint terms
| $0.840 | $2.64 | undeclared | 99.43% |
| Morphmorph Endpoint terms
| $0.869 | $2.73 | undeclared | 98.86% |
| InferenceNetinference-net/fp4 Endpoint terms
| $0.900 | $3.00 | fp4 | 98.52% |
| DigitalOceandigitalocean Endpoint terms
| $0.910 | $2.86 | undeclared | 99.57% |
| GMICloudgmicloud/fp8 Endpoint terms
| $0.980 | $3.08 | fp8 | 99.27% |
| Makoramakora/fp4 Endpoint terms
| $1.05 | $4.20 | fp4 | 97.65% |
| SiliconFlowsiliconflow/fp8 Endpoint terms
| $1.12 | $3.52 | fp8 | 99.98% |
| Decartdecart/fp4 Endpoint terms
| $1.19 | $3.74 | fp4 | 99.41% |
| Alibabaalibaba Endpoint terms
| $1.19 | $3.74 | undeclared | 97.74% |
| Waferwafer Endpoint terms
| $1.19 | $4.40 | undeclared | 99.88% |
| Sail Researchsail-research/us Endpoint terms
| $1.21 | $3.87 | fp8 | 99.78% |
| Friendlifriendli Endpoint terms
| $1.26 | $3.96 | undeclared | 100.00% |
| AkashMLakashml/fp8 Endpoint terms
| $1.30 | $4.40 | fp8 | 99.95% |
| Parasailparasail/fp8 Endpoint terms
| $1.40 | $4.40 | fp8 | 99.34% |
| AtlasCloudatlas-cloud/fp8 Endpoint terms
| $1.40 | $4.40 | fp8 | 98.68% |
| Z.AIz-ai/fp8 Endpoint terms
| $1.40 | $4.40 | fp8 | 99.89% |
| BaseTenbaseten/fp4 Endpoint terms
| $1.40 | $4.40 | fp4 | 99.84% |
| Crusoecrusoe/fp4 Endpoint terms
| $1.40 | $4.40 | fp4 | 99.54% |
| BaseTenbaseten/fp4 Endpoint terms
| $1.40 | $4.40 | fp4 | 99.71% |
| Mistralmistral Endpoint terms
| $1.40 | $4.40 | undeclared | 99.96% |
| PrimeIntellectprimeintellect Endpoint terms
| $1.40 | $4.40 | undeclared | 100.00% |
| Venicevenice Endpoint terms
| $1.40 | $4.40 | undeclared | 98.06% |
| Togethertogether Endpoint terms
| $1.40 | $4.40 | undeclared | 98.97% |
| Modalmodal Endpoint terms
| $1.40 | $4.40 | undeclared | 99.55% |
| Fireworksfireworks Endpoint terms
| $1.40 | $4.40 | undeclared | 99.51% |
| Cloudflarecloudflare Endpoint terms
| $1.40 | $4.40 | undeclared | 93.74% |
| BaseTenbaseten/fp8 Endpoint terms
| $2.10 | $6.60 | fp8 | 67.68% |
| BaseTenbaseten/fp8 Endpoint terms
| $2.10 | $6.60 | fp8 | 82.87% |
Independent scores
| LMArena Elo | 1475.1max |
|---|
LMArena Elo under CC BY 4.0. Absence of another board is not a score of zero.
Best rankings by task
- gamedev #3of 107 1367
- 3d #4of 100 1387
- codecategories #6of 108 1322
- svg #6of 76 1315
- uicomponent #7of 104 1341
- python-pptxslides #7of 26 1258
- website #8of 114 1309
- htmlslides #9of 23 1189
Capability
| Context window | 1.3M |
|---|---|
| Max output | 131K |
| Input modes | text |
| Tool use | yes |
| Reasoning | always on |
| Open weights | no |
Provenance
| Price source | openrouter.ai |
|---|---|
| Fetched | 2026-09-23 |
| Quality data | verified |
| Cross-checked | vendor page |
Our take
editorial — not a measurementGLM is a candidate for text-only workloads where selected-workload cost and task quality can be evaluated together. Use the current proof and benchmark evidence to judge measured price and capability. Do not infer downloadable weights or licence rights from the broader GLM family; verify the exact release and endpoint.
Strengths
- Text input, tools and reasoning effort are listed
- Cached-input pricing is recorded
Weaknesses
- Family history does not establish this release’s weight or licence status
- Text-only input is a migration constraint
Reach for it when
- Text-only workload evaluations
- Teams that compare cost per completed task
Avoid it if
- You require multimodal input
- Self-hosting is required but the exact weights and licence are not established
Sources: openrouter.ai
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...
Questions this page answers
What does GLM 5.3 cost?
$0.840/M in, $2.64/M out in this catalogue, as of the fetch date on this page. That is the API row, not a subscription.
Does GLM 5.3 have an independent quality score?
GLM 5.3 has an LMArena score in this catalogue.
What beats GLM 5.3?
Nothing is both better and cheaper than GLM 5.3.
Does this page use Artificial Analysis scores for GLM 5.3?
No. Artificial Analysis figures are not published here. Quality on this page is LMArena Elo where a score exists; otherwise the row is unrated.
Is the cheapest GLM 5.3 endpoint the same product?
Headline cheapest is a lower precision. Like-for-like at the best declared precision is $0.561/M from Baidu.
MODEL MONUMENT
Save or share this proof
Badge
The verdict as one SVG, for a README or a docs page. It states this model's dominance status at the balanced workload on the LMArena lens, with the date it was computed, and it is rebuilt with the catalogue — so it changes when the verdict changes, including to one you would rather it did not.
Markdown
[](https://undominated.ai/models/z-ai__glm-5.3/)HTML
<a href="https://undominated.ai/models/z-ai__glm-5.3/"><img src="https://undominated.ai/badge/z-ai__glm-5.3.svg" alt="GLM 5.3 — Undominated.ai dominance verdict" height="36"></a>Direct file: https://undominated.ai/badge/z-ai__glm-5.3.svg — a static SVG, written by scripts/build-badges.mjs on every build, so the copy you embed is never
older than the last deploy.
