Documentation Navigation

Models & pricing / Models / DeepSeek-V4-Pro-0813

DeepSeek-V4-Pro-0813 Overview

This model's data has not been fully verified and benchmarks may be missing.

DeepSeek-V4-Pro-0813 is a chat-reasoning AI model created by deepseek-ai, featuring 1.7T parameters and a context window of unknown.

DeepSeek-V4-Pro-0813 is an official release by DeepSeek-AI, built upon the DeepSeek-V4-Pro (Preview) model structure and enhanced with a DSpark speculative decoding module. It features greatly improved agentic capabilities and performance for production environments, supporting a `reasoning_effort` parameter for deliberation control. The model is designed for agentic scenarios and can be run with vLLM or SGLang.

Model Lineage & Specification

deepseek-v4-pro-0813 is deepseek-ai's primary release in the current family.

Developerdeepseek-ai
API Identifierdeepseek-v4-pro-0813
Parameters1.7T
Context Windowunknown
LicenseUnknown

Comparable models

DeepSeek-V4-Pro-0813

DeepSeek-V4-Pro-0813 is an official release by DeepSeek-AI, built upon the DeepS...

API IDdeepseek-v4-pro-0813
Typeopen weights
Parameters1.7T
Contextunknown

Solar Pro 4

Solar Pro 4 is a large language model from Upstage built for enterprise workflow...

API IDupstage-solar-pro-4
Typeapi only
Parameters
Context524288

Qwen3.8 Max Preview

Preview Qwen flagship for million-token multimodal reasoning and long-horizon ag...

API IDalibaba-qwen3.8-max-preview
Typeapi only
Parameters
Context1M tokens

Nemotron 3 Ultra

NVIDIA Nemotron 3 Ultra is a flagship open-weight frontier LLM family engineered...

API IDnvidia-nemotron-3-ultra
Typeopen weights
Parameters550B total / 55B active
Context1M tokens

DeepSeek-V4-Pro-0813 is an official release by DeepSeek-AI, built upon the DeepSeek-V4-Pro (Preview) model structure and enhanced with a DSpark speculative decoding module. It features greatly improved agentic capabilities and performance for production environments, supporting a `reasoning_effort` parameter for deliberation control. The model is designed for agentic scenarios and can be run with vLLM or SGLang.