Skip to main content
MindStudio
Pricing
BlogAbout
My Workspace
Text Generation Model

DeepSeek V4 Pro

DeepSeek V4 Pro is an MoE model with 1.6T total parameters (49B active) and a 1M-token context window. It's built for advanced reasoning, coding, and long-running agent tasks, and performs well on knowledge, math, and software engineering benchmarks.

PublisherDeepSeek
TypeText
Context Window1,000,000 tokens
Training DataApril 2026
Input$1.74/MTok
Output$3.48/MTok
ProviderDeepInfra
FLAGSHIPLATESTREASONING

DeepSeek V4 Pro

DeepSeek V4 Pro is an MoE model with 1.6T total parameters (49B active) and a 1M-token context window. It's built for advanced reasoning, coding, and long-running agent tasks, and performs well on knowledge, math, and software engineering benchmarks.

Ready to build with DeepSeek V4 Pro?

Get Started Free

Documentation & links

Parameters & options

Max Temperature1
Max Response Size384,000 tokens
Reasoning EffortSelect

Non-think for fast responses, High for complex problem-solving, Max to push reasoning to its fullest extent.

Default: high
Non-thinkLowMediumHighExtra High
Top PNumber

Nucleus sampling. Considers only tokens whose cumulative probability exceeds this threshold.

Default: 0.9Range: 0–1 (step 0.01)
Top KNumber

Limits sampling to the K most likely tokens at each step. Set to 0 to disable.

0Range: 0–100
Min PNumber

Minimum probability threshold relative to the most likely token.

0Range: 0–1 (step 0.01)
Presence PenaltyNumber

Penalizes tokens that have already appeared in the output, encouraging new topics.

0Range: -2–2 (step 0.01)
Frequency PenaltyNumber

Penalizes tokens based on how often they have already appeared.

0Range: -2–2 (step 0.01)
Repetition PenaltyNumber

Penalizes repeated tokens. Values above 1 discourage repetition.

Default: 1Range: 0–2 (step 0.01)
SeedSeed
Range: -1–2147483647

Start building with DeepSeek V4 Pro

No API keys required. Create AI-powered workflows with DeepSeek V4 Pro in minutes — free.