GLM-5.3
Intelligence
44.86
#35 of 520 models
Output speed
67.6 tok/s
Measured generation rate
Cost / index task
$2.006
Measured evaluation workload
End-to-end latency
40ms
First answer 33ms
01 / Position
A model in the landscape
Compare this model with peers across intelligence, speed, cost, and individual evaluations.
02 / Evidence
Benchmark results
23 distinct evaluations grouped by topic. Explore cross-model charts or switch to the full results list.
Domain capability
03 / Operation
Speed, tokens & price
Prices are per million tokens; latency and task costs describe the captured evaluation rather than a universal provider guarantee.
Token pricing
- Input / 1M
- $1.40
- Output / 1M
- $4.40
- Cache read / 1M
- $0.260
- Cache write / 1M
- —
- Cache-hit discount
- 81.4%
Response & usage
- Output speed
- 67.6 tok/s
- First chunk
- 3ms
- First answer
- 33ms
- Output / task
- 71.1k
- Reasoning / task
- 48.7k
Evaluation workload
- Canonical input
- 5.41B
- Canonical output
- 208.64M
- Answer / task
- 22.4k
- Index time / task
- 13.6m
- Run cost
- $2,503.479
By prompt type
| Prompt | Output speed | First chunk | First answer | End to end |
|---|---|---|---|---|
| Medium | 67.8 tok/s | 3ms | 32ms | 40ms |
| Long | 67.6 tok/s | 3ms | 33ms | 40ms |
| Hundred K | 78 tok/s | 8ms | 33ms | 40ms |
| Medium Parallel | 52.6 tok/s | 3ms | 42ms | 51ms |
04 / Access
License & availability
Recorded license
GLM-5.3 License
The publisher-linked license text defines permitted use and redistribution. Review it before using the model.
- Weight access
- Open weights
- Classification
- Commercial License
- Reasoning
- Yes
Routing and providers
- OpenRouter ID
- z-ai/glm-5.3
- Tokenizer
- Other
- Instruct type
- —
- Captured endpoints
- 34
| Provider | Quantization | Context | tok/s |
|---|---|---|---|
| Baidu | fp8 | 1.05M | — |
| Io Net | fp8 | 262.1k | — |
| Morph | unknown | 1.05M | — |
| Novita | fp8 | 1.05M | — |
| InferenceNet | fp4 | 1M | — |
| DeepInfra | fp4 | 1.05M | — |
| Phala | unknown | 1.05M | — |
| DigitalOcean | unknown | 1.05M | — |
| Reka | fp8 | 262.1k | — |
| Sail Research | fp8 | 1.05M | — |
| Inceptron | fp4 | 1.05M | — |
| GMICloud | fp8 | 1.05M | — |
| Makora | fp4 | 980k | — |
| SiliconFlow | fp8 | 1.05M | — |
| Alibaba | unknown | 1M | — |
| Decart | fp4 | 1.05M | — |
| Sail Research | fp8 | 1.05M | — |
| Friendli | unknown | 1.05M | — |
| AkashML | fp8 | 1.05M | — |
| BaseTen | fp4 | 1.05M | — |
| Mistral | nvfp4 | 1.05M | — |
| Crusoe | fp4 | 1.05M | — |
| Wafer | unknown | 1.05M | — |
| Venice | unknown | 1M | — |
| Together | unknown | 1.05M | — |
| Parasail | fp8 | 1.05M | — |
| Modal | unknown | 1.05M | — |
| BaseTen | fp4 | 1.05M | — |
| Fireworks | unknown | 1.05M | — |
| Cloudflare | unknown | 1.31M | — |
| AtlasCloud | fp8 | 1.05M | — |
| Z.AI | fp8 | 1.05M | — |
| BaseTen | fp8 | 1.05M | — |
| BaseTen | fp8 | 1.05M | — |
05 / Audit
Underlying fields
Inspect the captured model and routing records. Artificial Analysis destination links are omitted.