DeepSeek Releases V4 Pro: 1.6T Parameter MoE Model with 1M Token Context at $1.74/M Input Tokens
DeepSeek has released V4 Pro, a Mixture-of-Experts model with 1.6 trillion total parameters and 49 billion activated parameters. The model supports a 1-million-token context window and costs $1.74 per million input tokens and $3.48 per million output tokens.
DeepSeek V4 Pro — Quick Specs
DeepSeek Releases V4 Pro: 1.6T Parameter MoE Model with 1M Token Context
DeepSeek has released V4 Pro, a large-scale Mixture-of-Experts model with 1.6 trillion total parameters and 49 billion activated parameters, supporting a 1-million-token context window. The model is priced at $1.74 per million input tokens and $3.48 per million output tokens.
Architecture and Capabilities
According to DeepSeek, V4 Pro is built on the same architecture as DeepSeek V4 Flash and introduces a hybrid attention system designed for efficient long-context processing. The model supports multiple reasoning modes that allow users to balance speed and depth depending on task requirements.
The company claims the model delivers strong performance across knowledge, mathematics, and software engineering benchmarks, though specific benchmark scores have not been disclosed.
Target Use Cases
DeepSeek positions V4 Pro for complex workloads including:
- Full-codebase analysis
- Multi-step automation
- Large-scale information synthesis
- Advanced reasoning tasks
- Long-horizon agent workflows
The 1-million-token context window enables processing of entire codebases or lengthy documents in a single inference call.
Pricing and Availability
V4 Pro is available through OpenRouter as of April 24, 2026. At $1.74 per million input tokens, it sits in the mid-range pricing tier for frontier models. The 2:1 output-to-input pricing ratio ($3.48 vs $1.74) is standard for models with generation-heavy workloads.
Technical Details
The Mixture-of-Experts architecture activates 49 billion parameters per forward pass while maintaining 1.6 trillion total parameters. This approach aims to provide capabilities comparable to dense models of similar active parameter count while reducing computational costs.
OpenRouter integration includes support for DeepSeek's reasoning modes, with developers able to access step-by-step thinking processes through the reasoning_details array in API responses.
What This Means
V4 Pro represents DeepSeek's entry into the ultra-long-context market dominated by models like Anthropic's Claude and Google's Gemini. The 1M token context window and competitive pricing make it viable for enterprise use cases requiring analysis of large documents or codebases. However, without published benchmark scores, direct performance comparisons to established models remain unclear. The MoE architecture suggests DeepSeek is prioritizing inference efficiency alongside capability, a trend across Chinese AI labs competing with Western frontier model providers.
Related Articles
DeepSeek V4.1 Flash 1017 launches with 552B CED backbone, 1M context, $0.15/$0.60 per 1M tokens
DeepSeek has released V4.1 Flash 1017, an October 17, 2026 snapshot built on its new Causal Encoder-Decoder architecture. It uses a 552B-parameter sparse MoE backbone with a 1M-token context window, native image understanding, and pricing of $0.15 input / $0.60 output per 1M tokens on OpenRouter.
Recursive releases Flint, a 40B-parameter open coding model that runs on laptops under Apache 2.0
Recursive released Flint, a free, open-weight mixture-of-experts coding model with 40 billion total and 2.5 billion active parameters, under Apache 2.0. The company claims its Eureka Machine automated research system contributed to nearly every stage of development, and that Flint runs at about 131 tokens per second on an M4 Pro MacBook Pro.
Deepseek open-sources DeepInfer, the Apache 2.0 inference engine behind its V4.1-Flash model
Deepseek has open-sourced DeepInfer, the internal inference framework that serves its V4.1-Flash model, under the Apache License 2.0. The company says it will also open-source code for future models and has released deepseek-recipe to reproduce its API's token processing in self-hosted setups.
MiniMax M3.1 Flash scores 42 on AA Intelligence Index at $0.15/$0.60 per 1M tokens, open weights Oct 23
MiniMax has released M3.1 Flash, a reasoning model that scores 42 on the Artificial Analysis Intelligence Index, up from 29 for M3. It costs $0.15 per 1M input tokens and $0.60 per 1M output tokens, and open weights are planned for October 23.
Comments
Loading...