NVIDIA NIM

NVIDIA NIM: DeepSeek V4 Pro (NIM)

stable

DeepSeek V4 Pro (NIM) is available through Everstack's OpenAI-compatible gateway with routing, fallback, retries, and observability.

deepseek-ai__deepseek-v4-prodeepseek-v4released Apr 24, 2026
ProviderQuick start
Context
1.0M
tokens
Max output
393.2K
tokens
Input
$0.43
per 1M
Output
$0.87
per 1M

Activity

Daily period usage
Aggregated managed Everstack traffic. Buckets below 5 tenants or 100 requests are withheld.

Providers

ProviderContextInput /MOutput /M
DeepSeek
DeepSeek
1M$0.43$0.87
Hugging Face
Hugging Face
1.0M$0.43$0.87
NVIDIA NIM
NVIDIA NIM
1.0M$0.43$0.87
OpenRouter
OpenRouter
1.0M$0.57$1.15
Fireworks AI
Fireworks AI
1M$1.74$3.48
Together AI
Together AI
512K$1.74$3.48

Pricing

ItemPriceUnit
Input$0.43per 1M tokens
Output$0.87per 1M tokens
Cache read$0.004per 1M tokens

Performance

Reliability

Compare

Compare usage and performance
Place this model beside up to three other catalog models.

Capabilities

Capabilities
Modalities
Input
Output
Details
Model ID
deepseek-ai__deepseek-v4-pro
Family
deepseek-v4
Status
stable
Knowledge
-
Released
Apr 24, 2026
Gateway Config
API
2024-01-01
Auth
NVIDIA_API_KEY
Discovery
static
Provider type
meta

Quick Start

index.ts
import Everstack from "@everstack/node";const client = new Everstack({  apiKey: process.env.EVERSTACK_API_KEY!,  baseUrl: "https://{instance}.{region}.everstack.ai",});const completion = await client.chat.completions.create({  model: "@nvidia-nim/deepseek-ai__deepseek-v4-pro",  messages: [{ role: "user", content: "Explain vector cache routing." }],});console.log(completion.choices[0]?.message?.content);

Endpoint

Base URL
https://integrate.api.nvidia.com/v1
Chat completions
https://integrate.api.nvidia.com/v1/chat/completions
Model
deepseek-ai__deepseek-v4-pro
Auth env
NVIDIA_API_KEY

DeepSeek V4 Pro (NIM) FAQ

How much does the DeepSeek V4 Pro (NIM) API cost?

Through NVIDIA NIM, DeepSeek V4 Pro (NIM) costs $0.43 per 1M input tokens and $0.87 per 1M output tokens.

What is the context window of DeepSeek V4 Pro (NIM)?

DeepSeek V4 Pro (NIM) accepts up to 1.0M tokens of context and returns up to 393.2K output tokens on NVIDIA NIM.

Which providers offer DeepSeek V4 Pro (NIM)?

DeepSeek V4 Pro (NIM) is available from DeepSeek, Hugging Face, NVIDIA NIM, OpenRouter, Fireworks AI and Together AI. Everstack can route between them with automatic fallback when one is unavailable.

What can DeepSeek V4 Pro (NIM) do?

DeepSeek V4 Pro (NIM) supports chat, function calling and reasoning. It accepts text input.

How do I call DeepSeek V4 Pro (NIM) through Everstack?

Point any OpenAI-compatible SDK at your Everstack gateway and set the model to @nvidia-nim/deepseek-ai__deepseek-v4-pro. Everstack adds retries, provider fallback, caching, per-tenant budgets and OpenTelemetry traces for every request.