DeepSeek V4 Pro 0813

About this model

DeepSeek V4 Pro 0813 is the official release of DeepSeek V4 Pro, a 1.6T parameter Mixture-of-Experts (MoE) chat model from DeepSeek AI with an attached DSpark speculative decoding module for accelerated inference. It supports a one-million-token context window and three reasoning effort levels (low, high, max) for controllable deliberation, with up to 384K output tokens recommended at high and max levels. The model achieves 60.0 on HLE (with tools), 87.9 on Terminal Bench 2.1, and 74.1 on Toolathlon-Verified, demonstrating strong agentic and coding-agent capabilities. It is released under the MIT License.

Technical specifications

Capabilities

Input modalities

Output modalities

Reasoning

Hybrid Default on

Knowledge horizon