Model
Model
z-ai
GLM 5.2 is a large-scale reasoning model from Z.ai. It supports text input and output with a 1M-token context window, and is suited for long-horizon agent workflows, project-level software engineering,...
| Provider | Quantization | Input / M tokens | Output / M tokens |
|---|---|---|---|
| Waferwafer/us | fp4 | $0.0341 | $7.00 |
| Cloudflarecloudflare | — | $0.19 | $8.00 |
| Waferwafer | fp4 | $0.19 | $8.00 |
| Morphmorph/fp8 | fp8 | $0.207 | $4.40 |
| Relacerelace | — | $0.40 | $4.00 |
| DeepInfradeepinfra/fp4/flex | fp4 | $0.45 | $1.44 |
| DeepInfradeepinfra/fp4 | fp4 | $0.5625 | $1.80 |
| StreamLakestreamlake/fp8 | fp8 | $0.61306 | $1.92676 |
| CoreWeavecoreweave/fp4 | fp4 | $0.62 | $2.42 |
| Novitanovita/fp8 | fp8 | $0.6496 | $2.0416 |
| Inceptroninceptron/fp4 | fp4 | $0.69 | $2.52 |
| DigitalOceandigitalocean | — | $0.70 | $2.20 |
| SiliconFlowsiliconflow/fp8 | fp8 | $0.70 | $2.20 |
| AtlasCloudatlas-cloud/fp8 | fp8 | $0.938 | $2.948 |
| Alibabaalibaba/fp8 | fp8 | $0.966 | $3.036 |
| Alibabaalibaba/fp8/us-east-1 | fp8 | $1.12 | $3.52 |
| Decartdecart/fast | fp4 | $1.20 | $8.00 |
| Phalaphala/fp8 | fp8 | $1.26 | $3.00 |
| Baidubaidu/fp8 | fp8 | $1.40 | $4.40 |
| BaseTenbaseten/fp8 | fp8 | $1.40 | $4.40 |
| BaseTenbaseten/fp8/us | fp8 | $1.40 | $4.40 |
| Friendlifriendli | — | $1.40 | $4.40 |
| GMICloudgmicloud/fp8 | fp8 | $1.40 | $4.40 |
| Mistralmistral/zdr | nvfp4 | $1.40 | $4.40 |
| Nebiusnebius/fp4 | fp4 | $1.40 | $4.40 |
| Parasailparasail/fp4 | fp4 | $1.40 | $4.40 |
| Togethertogether | — | $1.40 | $4.40 |
| Venicevenice/fp8 | fp8 | $1.40 | $4.40 |
| Z.AIz-ai/fp8 | fp8 | $1.40 | $4.40 |
| Mistralmistral/eu | — | $1.54 | $4.84 |
| BaseTenbaseten/fast | fp8 | $2.10 | $6.60 |
| BaseTenbaseten/us/fast | fp8 | $2.10 | $6.60 |
| Baidubaidu/fast | fp4 | $2.25 | $7.88 |
| Alibabaalibaba/fast | fp8 | $2.31 | $7.26 |
Prices are what the Provider charges, synced 2026-10-21. Amber adds its platform fee to each Trial’s total; see pricing.
Amber has no curated Model Configuration for this Model yet. Create a custom one in your Workspace to evaluate it.