DeepSeek·DeepSeek V family

DeepSeek V4

Open WeightsMoEmitReleased Feb 2026

DeepSeek's fourth-generation model with improved mixture-of-experts architecture, enhanced reasoning and coding capabilities, and stronger multilingual performance. Competitive with frontier proprietary models.

Capabilities

chatcompletionfunction-callingcode-generationreasoning

Modalities

textcode

Context Window

256K tokens

Providers

available

Available from 6 providers

Cheapest

SiliconFlow

$0.42/1M tokens

Anyscale, DeepSeek, Fireworks, Hugging Face Inference, SiliconFlow, Together AI

Providers (6)

Sorted by total cost (input + output per 1M tokens). Click a row to view provider details.

Provider	Pricing (per 1M)	Rate Limits	Regions	Health	Latency
SiliconFlow	In: $0.14Out: $0.28	600 RPM / 1.0M TPM	ap-east-1global	Healthy	0ms
Hugging Face Inference	In: $0.18Out: $0.36	300 RPM / 500K TPM	us-east-1eu-west-1	Unhealthy	0ms
Anyscale	In: $0.20Out: $0.40	300 RPM / 500K TPM	us-east-1us-west-2	Healthy	0ms
Fireworks	In: $0.20Out: $0.80	600 RPM / 1.0M TPM	us-east-1us-west-2	Healthy	0ms
DeepSeek	In: $0.27Out: $1.10	500 RPM / 1.0M TPM	us-east-1global	Healthy	0ms
Together AI	In: $0.90Out: $0.90	600 RPM / 1.0M TPM	us-east-1us-west-2	Healthy	0ms

Quick Start

Use this model via SiliconFlow with an OpenAI-compatible SDK.

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.siliconflow.com/v1",
  apiKey: process.env.SILICONFLOW_API_KEY,
});

const response = await client.chat.completions.create({
  model: "deepseek-ai/DeepSeek-V4",
  messages: [
    { role: "user", content: "Hello!" }
  ],
});

console.log(response.choices[0].message.content);

Using SiliconFlow API • OpenAI-compatible SDK