Model
Model
deepseek
DeepSeek V4 Flash 0731 is a sparse mixture-of-experts model from DeepSeek, with 13B active parameters out of 284B total. This re-post-trained revision is suited for coding, reasoning, and agent workflows....
| Provider | Quantization | Input / M tokens | Output / M tokens |
|---|---|---|---|
| Relacerelace/fp4 | fp4 | $0.0475 | $0.1568 |
| Rekareka | — | $0.05 | $0.528 |
| DeepInfradeepinfra/fp8 | fp8 | $0.06 | $0.18 |
| Inceptroninceptron/fp4 | fp4 | $0.11 | $0.4125 |
| DigitalOceandigitalocean | — | $0.119 | $0.238 |
| BaseTenbaseten/fp32 | fp32 | $0.13 | $0.26 |
| BaseTenbaseten/fp8 | fp8 | $0.13 | $0.26 |
| Venicevenice | — | $0.13 | $0.26 |
| CoreWeavecoreweave/fp8 | fp8 | $0.13 | $0.28 |
| StreamLakestreamlake/fp8 | fp8 | $0.132 | $0.396 |
| Waferwafer/fast | nvfp4 | $0.14 | $0.23 |
| Coherecohere | — | $0.14 | $0.28 |
| Parasailparasail/fp8 | fp8 | $0.14 | $0.28 |
| Togethertogether | — | $0.14 | $0.28 |
| Sail Researchsail-research/fp4 | fp4 | $0.15 | $0.15 |
| Sail Researchsail-research/us | fp4 | $0.15 | $0.15 |
| Mancer 2mancer/fp8 | fp8 | $0.20 | $0.60 |
| SiliconFlowsiliconflow/fp8 | fp8 | $0.22 | $0.66 |
| GMICloudgmicloud/fp8 | fp8 | $0.286 | $0.858 |
| Phalaphala | — | $0.308 | $0.924 |
| Alibabaalibaba | — | $0.352 | $1.056 |
| Alibabaalibaba/us-east-1 | — | $0.374 | $1.122 |
| Novitanovita/fp8 | fp8 | $0.4092 | $1.2276 |
| AtlasCloudatlas-cloud/fp4 | fp4 | $0.44 | $1.32 |
| Baidubaidu/fp8 | fp8 | $0.44 | $1.32 |
| Cloudflarecloudflare | — | $0.44 | $1.32 |
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.