GLM-5.3
Z.ai's large open model. Near Qwen3.8 Max on the index, with a lower measured task cost.
Open weights
45Index · Max
$2.01Per measured task
22.4 pts/$Points / $
68Tokens / sec
No High score is published. Showing Max.
Every published effort
| Setting | Index | Per task | Tokens / sec |
|---|---|---|---|
| Max | 45 | $2.01 | 68 |
| Low | 34 | $0.85 | 69 |
What this workload costs
List price for one call. Reasoning tokens count as output. A cache hit uses the lab’s cache rate when one is published.
GLM-5.3
No cache priceSpecs
- API id
- Not published here
- Released
- Sep 2026
- Context
- 1M
- Max output
- —
- Input / 1M
- $1.40
- Output / 1M
- $4.40
- Cache read / 1M
- —
- Knowledge
- —
- Inputs
- Text
- Weights
- GLM-5.3 license, not MIT. Read it before you embed the weights.
Vendor list price as compiled on 23 Sep 2026. A yuan-denominated bill can print a slightly different USD figure.
Vision is confirmed below for the Flash model, not for this one, in the sources used here.
Use it when
Open-weight long-horizon coding when you want the larger GLM. Index 45 at max, $2.01 per measured task.
Leave it when
If MiMo-V2.6-Pro's license fits the product. It scores a point higher on this index and costs much less per measured task.