Fireworks AI Hosts DeepSeek V4 Pro With 1M Context and MIT License

Fireworks AIFireworks AI

· Updated

Fireworks AI added DeepSeek V4-Pro to its inference platform, offering frontier-level coding and reasoning via an open-weight model. The deployment standardizes a 1-million token context window at a price point significantly lower than closed-source competitors.

Fireworks AI, an inference platform for fast model serving, launched support for DeepSeek V4-Pro on its cloud. This flagship Mixture-of-Experts (an architecture activating only a fraction of parameters per request) model is now available with a standard 1-million token context window, enabling developers to process massive datasets.
Model
DeepSeek V4-Pro
Context window
1 million tokens
License
MIT
Pricing (input)
$1.74 per million tokens
Pricing (output)
$3.48 per million tokens

The release follows the initial DeepSeek V4 preview and marks a shift toward making high-end reasoning models economically viable. By providing Day-0 support for frontier models, the platform allows teams to swap closed-source APIs for open-weight alternatives that match top-tier coding performance without restrictive licensing.

You can access the model via API for $1.74 per million input tokens and $3.48 per million output tokens. The MIT license enables full commercial use for long-context tasks like codebase analysis. This follows the platform's recent expansion into distributed training infrastructure for large-scale agentic systems.

View the full update on app.fireworks.ai
Fireworks AI
Fireworks AI
@FireworksAI_HQ
X

DeepSeek V4-Pro is live on Fireworks: → Frontier-class coding and reasoning, open-weights → 1M context standard, MIT licensed → $1.74 / $3.48 per 1M tok Get started now → https://t.co/BKrdxniuyR https://t.co/T6NGGWgHme

9retweets189likes
View on X

Still wondering? A few quick answers below.

DeepSeek V4-Pro is a frontier-class large language model featuring a Mixture-of-Experts architecture. It is specifically optimized for high-performance coding and complex reasoning tasks, rivaling leading closed-source models in these technical domains. The model is designed to handle sophisticated logic and programming challenges with high efficiency.

Fireworks AI provides a standard 1-million token context window for DeepSeek V4-Pro. This massive window allows the model to process and reason across extremely large datasets, such as entire software codebases or lengthy technical documents, in a single interaction. This capability is essential for agentic workflows that require deep understanding.

DeepSeek V4-Pro is an open-weight model released under the MIT license, which is one of the most permissive licenses available. This allows developers and enterprises to use, modify, and distribute the model for commercial purposes without the restrictive terms often found in proprietary AI systems, providing more flexibility for production.

Fireworks AI offers DeepSeek V4-Pro at a competitive price point of $1.74 per 1 million input tokens and $3.48 per 1 million output tokens. This pricing structure makes frontier-level reasoning and coding capabilities significantly more accessible compared to closed-source alternatives, enabling developers to build and scale complex AI applications.

Every HeadsUpAI update is written based on its original source and reviewed before it's published. Read our editorial standards →

Fireworks AI

See all AI news & updates from Fireworks AI →

Share this update