Back to all models

DeepSeek V4 Pro (W&B Inference)

DeepSeek V4-Pro served on W&B Inference (CoreWeave). Stronger reasoning and coding than V4-Flash at $1.74/$3.46 per 1M tokens — a solid step up for tasks that need more than the cheap tier but don't require a frontier proprietary model.

wandb-deepseek-v4-pro
STABLEGet Started
1,048,576 context
Starting at $1.74/M input tokens
Starting at $3.46/M output tokens
Streaming
Tools
Reasoning
JSON Output

Select Provider

All Providers for DeepSeek V4 Pro (W&B Inference)

LLM Gateway routes requests to the best providers that are able to handle your prompt size and parameters.

W&B Inference (CoreWeave)
Context: 1.0M
Input
$1.74
/M tokens
Cached
$0.14
/M tokens
Output
$3.46
/M tokens
Get Started