Key Highlights
- V4-Pro model receives promotional 75% price reduction through May 5, 2026
- API input cache hit pricing slashed by 90% across all models effective immediately
- Two model variants available: full-featured Pro and streamlined Flash edition
- Model engineered for compatibility with Huawei semiconductor infrastructure, exceeding competing open-source alternatives in knowledge assessments
- Aggressive pricing strategy reflects escalating competition in the global artificial intelligence marketplace
Hangzhou-based artificial intelligence developer DeepSeek has announced dramatic pricing reductions for its latest V4-Pro model, implementing a 75% discount as competitive pressure intensifies between Chinese and international AI enterprises.
The promotional pricing structure for software developers went into effect last week and will remain available until 15:59 UTC on May 5, 2026.
With these adjustments, input costs for cache misses decreased from $1.74 to $0.435. Cache hit pricing fell from $0.145 to $0.03625, while output costs dropped from $3.48 to $0.87 per unit.
In addition to the V4-Pro discount, DeepSeek implemented a 90% reduction on input cache hit pricing throughout its complete API portfolio. According to the company, this permanent pricing adjustment began immediately and will benefit developers who submit recurring or similar API requests.
The V4-Pro model’s debut follows an extended development period. Engineers optimized the system for Huawei’s semiconductor technology—a significant consideration given that American export controls have restricted Chinese firms’ ability to procure U.S.-manufactured chips.
Dual Model Strategy
DeepSeek’s V4 release includes two distinct configurations. The Pro configuration delivers enhanced capabilities and carried premium pricing before the discount implementation. The Flash configuration offers a more efficient, cost-effective alternative.
According to DeepSeek’s performance data, the Pro configuration surpasses competing open-source models in global knowledge evaluation benchmarks. Only Google’s proprietary Gemini-Pro-3.1 system achieves superior results in these assessments.
The company positions the V4 models as optimized for AI agent applications. These sophisticated systems can execute more intricate operations than conventional chatbot interfaces, though they demand greater computational resources.
This pricing initiative follows DeepSeek’s earlier R1 model release, which sparked widespread price competition throughout the AI sector upon its introduction last year.
Industry-Wide Competitive Pricing
As artificial intelligence companies transition from experimental phases to commercial deployment of large language models, reducing operational and inference expenses has emerged as a critical competitive differentiator.
DeepSeek’s pricing adjustments are anticipated to compel competing firms to implement corresponding reductions, particularly within China’s market, where organizations are developing alternatives to Western technological platforms.
American technology export restrictions have accelerated this transformation, catalyzing the expansion of domestic AI infrastructure throughout China.
OpenAI, Anthropic, and Google continue launching new model iterations at an accelerating pace. Premium access to these platforms creates an opening for DeepSeek’s value-oriented pricing approach.
The 75% promotional discount on V4-Pro continues through May 5, while the comprehensive API pricing reductions across DeepSeek’s entire model range are currently in effect.
Get 3 Free Stock Ebooks
Discover top-performing stocks in AI, Crypto, and Technology with expert analysis.
- Top 10 AI Stocks - Leading AI companies
- Top 10 Crypto Stocks - Blockchain leaders
- Top 10 Tech Stocks - Tech giants
