DeepSeek

DeepSeek V4 Pro

deepseek
Chat

DeepSeek V4 Pro is a 1.6T-parameter Mixture-of-Experts model with 49B active parameters and a 1M-token context window. Built for advanced reasoning, coding, and long-horizon agentic workflows with a hybrid attention system for efficient long-context processing.

Modalities

In / out price

$1.65 / $3.301

per 1M

Context

1M

Max output

33K

Released

Apr 24, 2026

Updated

Jul 17, 2026

Privacy

1 / 3

Anonymous routing

Anonymous tier

  1. Anonymous
  2. Private
  3. E2EE
  • Your identity is hidden from the inference provider.
  • The inference provider can see prompt content; zero retention is not guaranteed.

Moderation

Not specified

No model-level moderation classification is recorded in this catalog.

Data retention

Not guaranteed

AnonRouter stores no payloads. Provider zero-retention is not guaranteed.

Features

7
Streaming
Tool calling
Reasoning
JSON/schema
Web search
Code optimized
Prompt caching

Routing

2

Anonrouter hosted

Routed through AnonRouter's gateway with metadata-only logging.

Provider direct

Requests egress directly to the provider runtime.

USD per 1M tokens from the Venice models API.

Providers & sources

1 route

Uptime

30d