deepseek-v4-flash-0731 · deepseek-v4-proBoth DeepSeek V4 Flash 0731 and DeepSeek V4 Pro are accessible through a single Newton’s API key using standard OpenAI SDKs.
Live per-million-token pricing served through Newton’s unified OpenAI-compatible gateway.
| Metric | DeepSeek: DeepSeek V4 Flash 0731 | DeepSeek: DeepSeek V4 Pro |
|---|---|---|
| API Model ID | deepseek-v4-flash-0731 | deepseek-v4-pro |
| Provider Lab | DeepSeek | DeepSeek |
| Gateway Status | Catalog | Catalog |
| Input / 1M tokens | — | — |
| Cached Input / 1M tokens | — | — |
| Output / 1M tokens | — | — |
| Context Window | 128K+ | 128K+ |
Estimated spend for typical coding and autonomous agent workloads based on published per-token rates:
Because Newton’s exposes both deepseek-v4-flash-0731 and deepseek-v4-pro behind the same API key and base URL, engineering teams frequently combine both models in the same workflow:
Point the official OpenAI SDK, Cursor, or Claude Code at https://newtons.dev/api/v1 and swap the model string on demand:
import OpenAI from "openai";
const client = new OpenAI({
apiKey: process.env.NEWTONS_API_KEY,
baseURL: "https://newtons.dev/api/v1",
});
// Switch between DeepSeek V4 Flash 0731 and DeepSeek V4 Pro with 1 line:
const response = await client.chat.completions.create({
model: "deepseek-v4-flash-0731", // or "deepseek-v4-pro"
messages: [{ role: "user", content: "Compare architecture trade-offs." }],
});