Chinese artificial intelligence startup DeepSeek has unveiled what researchers describe as the world’s most cost-efficient mainstream AI model, as China pushes its strategy of competing with U.S. technology leaders on affordability rather than raw computing power.

According to research firm Artificial Analysis, DeepSeek’s newly released V4-Flash model is significantly cheaper to operate than competing frontier models, with benchmark testing indicating it costs more than 100 times less to run than Anthropic’s Claude Fable 5 while delivering competitive performance across a range of reasoning and coding tasks.

The release comes as DeepSeek seeks to regain the spotlight in a crowded Chinese AI market and amid reports that the company is preparing for a potential initial public offering (IPO).

DeepSeek officially launched V4-Flash on Friday, extending the pricing strategy that first propelled the startup onto the global stage earlier this year. Its breakthrough R1 reasoning model shocked Silicon Valley in early 2025 by demonstrating that advanced AI systems could be developed at a fraction of the cost associated with leading U.S. models, triggering a sharp selloff in technology stocks and intensifying scrutiny of the hundreds of billions of dollars American companies have committed to AI infrastructure.

Artificial Analysis estimates that V4-Flash costs approximately $0.03 per benchmark test, making it the least expensive well-known AI model currently evaluated by the research firm.

That compares with an estimated $0.86 for Moonshot AI’s Kimi K3, $1.86 for OpenAI’s GPT-5.6 Sol and $3.15 for Anthropic’s Claude Fable 5. The pricing differential highlights how aggressively Chinese developers are competing on operating costs, a factor becoming more important as businesses move from experimenting with AI to deploying models at enterprise scale.

Pricing Alone Does Not Tell The Full Story

Artificial Analysis noted that benchmark cost provides a more meaningful comparison than headline token pricing because it measures the actual expense required to complete representative workloads. While some models advertise low token prices, they may require substantially more computational steps or generate longer responses, increasing total operating costs.

DeepSeek’s V4-Flash charges $0.14 per million input tokens and $0.28 per million output tokens, placing it among the industry’s cheapest commercially available frontier models.

That pricing could make the model particularly attractive to developers and enterprises deploying AI across high-volume customer service, coding assistance, and workflow automation applications, where inference costs often become one of the largest operational expenses.

Despite its aggressive pricing, V4-Flash delivers competitive benchmark results. Artificial Analysis awarded the model 50 points on its Intelligence Index, which combines results from nine standardized evaluations covering reasoning, coding, workplace productivity and general problem-solving tasks.

The score matches Google’s Gemini 3.6 Flash and trails Meta Platforms’ Muse Spark 1.1 and Zhipu AI’s GLM-5.2 by just one point.

However, more capable frontier systems continue to maintain a performance advantage.

Moonshot AI’s Kimi K3 achieved 57 points, while Anthropic’s Claude Opus 5, Claude Fable 5, and OpenAI’s GPT-5.6 scored at least nine points higher than DeepSeek’s latest release.

The results suggest DeepSeek continues to prioritize price-performance optimization rather than competing directly for the industry’s highest benchmark scores.

DeepSeek no longer dominates China’s AI landscape as decisively as it did after releasing R1. The company now faces fierce competition from domestic startups including Moonshot AI, MiniMax and Zhipu AI, as well as technology giants such as Alibaba Group and ByteDance, all of which are racing to capture global enterprise customers.

The competition now centers on lowering inference costs while maintaining acceptable performance, reflecting a broader shift across the AI industry toward commercialization and large-scale deployment rather than purely advancing benchmark performance.

The rivalry intensified further on Monday when Alibaba introduced Qwen3.8-Max, its largest and most powerful AI model to date, underscoring the rapid pace at which Chinese companies continue to iterate and release increasingly capable systems.

IPO Ambitions and Next-Generation Models

DeepSeek’s latest launch also comes as the company reportedly explores a public listing, a move that would provide additional capital to expand research, computing infrastructure and international operations.

Meanwhile, the startup is already preparing a more advanced model known as V4-Pro, although it has not announced an official release date. The staggered rollout suggests DeepSeek is pursuing a two-tier product strategy: highly affordable models aimed at broad commercial adoption alongside more powerful systems intended to compete with the most advanced offerings from U.S. rivals.

DeepSeek emerged as one of the most influential AI startups in 2025 after demonstrating that competitive large language models could be developed using significantly fewer computing resources than many Western counterparts. Its rapid rise challenged assumptions about the scale of investment required to build frontier AI and intensified competition between Chinese and American developers.

Cost efficiency is becoming nearly as important as raw model capability in the AI industry. As enterprises evaluate AI based on total deployment costs rather than benchmark performance alone, developers are competing to deliver the best balance between intelligence, speed and affordability, making operational efficiency a critical battleground in the global AI race.

SpaceX heads into its first earnings report as a publicly traded company under mounting pressure after losing more than $500 billion in market value since its June 12 stock market debut, with investors increasingly questioning whether its ambitious artificial intelligence and space expansion plans can justify its lofty valuation.

The Elon Musk-led company has fallen more than 50% from its intraday high and is coming off its fourth consecutive weekly decline, making it one of the weakest-performing high-profile technology IPOs in recent years.

The decline comes as Wall Street’s attitude toward AI investments has shifted markedly. After months of rewarding companies for aggressive spending on artificial intelligence infrastructure, investors are now demanding evidence that massive capital expenditure is translating into sustainable revenue growth, profitability and free cash flow.

That changing sentiment has already weighed on several AI leaders this earnings season and now puts the spotlight squarely on SpaceX, whose valuation depends far more on future execution than current financial performance. Unlike established technology giants generating billions of dollars in recurring AI-related profits, SpaceX is still investing heavily across launch services, satellite broadband, orbital computing infrastructure and artificial intelligence.

According to its IPO filing, the company continues to burn billions of dollars each quarter while carrying nearly twice as much debt as cash. Even after the selloff, SpaceX trades at a price-to-sales multiple in the 70s, reflecting investors’ expectations for extraordinary future growth.

Guidance Expected To Outweigh Quarterly Results

Analysts quoted by CNBC broadly expect Tuesday’s earnings report to be judged less on quarterly financial performance than on management’s outlook for growth, AI monetization and capital spending.

The report also comes just ahead of the expiration of rolling IPO lock-up restrictions, allowing early investors to begin selling shares, a development that could add further pressure to the stock.

Short sellers have already benefited substantially from the decline. Matthew Unterman, head of research at S3 Partners, described the bearish positioning as unusually aggressive.

“It’s among the most aggressive and quickest bearish builds we have seen in a mega-cap name heading into its first earnings report post-IPO,” Unterman said.

For many analysts, the company’s valuation ultimately depends on Starship, the fully reusable launch vehicle expected to dramatically reduce launch costs while enabling SpaceX to rapidly expand its Starlink satellite network and future orbital AI infrastructure.

The company itself underscored Starship’s importance in its IPO prospectus.

“If Starship does not achieve full reusability or rapid turnaround, we may experience higher per-launch costs, slower deployment timelines for our large-scale constellations (including our orbital AI compute program), delayed revenue growth, and increased overall capital requirements, and our brand and reputation may suffer,” SpaceX said in the filing.

The company also stated that it expects Starship to “commence payload delivery to orbit” during the second half of the year.

While the latest Starship test flight in July achieved several milestones, the Super Heavy booster experienced a hard splashdown after only some of its engines reignited during landing, highlighting that technical challenges remain before full operational capability is achieved.

Bernstein analysts said Starship remains the single biggest issue for investors trying to justify the company’s valuation.

Investors are also expected to seek greater clarity on SpaceX’s rapidly expanding artificial intelligence business following its merger with xAI earlier this year.

The company is attempting to build a new revenue stream around AI infrastructure, including hosted computing capacity, orbital data centers and AI services, while also developing its Grok chatbot and integrating the planned acquisition of AI coding startup Cursor.

SpaceX has already secured several high-profile commercial agreements.

Days before its IPO, the company signed a deal with Google expected to generate about $920 million in monthly revenue by providing AI computing capacity. It has also reached agreements with Anthropic and Reflection AI to supply compute infrastructure.

Those partnerships are important because they allow SpaceX to generate recurring infrastructure revenue while continuing to invest heavily in AI development.

Analysts at Cantor believe those businesses could become an important catalyst.

“We think earnings can meaningfully alleviate some of those pressures as the company proves out hosted-compute profits, clarifies capital funding geographies, and clears its initial lockup headwinds,” the analysts wrote.

“We view SPCX as approaching a bottom into the print.”

Some Analysts See The Selloff As An Opportunity

Not everyone on Wall Street believes the recent decline reflects deteriorating fundamentals. Ben Harwood, an analyst at New Street Research, noted that the selloff has created an attractive long-term entry point.

“For a long-term investor we think this is an attractive entry point,” Harwood wrote.

“The growth runway is enormous, and SpaceX has one of the widest moats in the market today.”

Bernstein analysts were similarly focused on management’s long-term execution rather than quarterly earnings.

“We believe the quarterly results should not matter,” the analysts said.

“What will be important is the level of confidence projected by management regarding the company’s growth path.”

SpaceX’s earnings arrive as investors become increasingly selective across the AI sector. Recent earnings from Alphabet, Meta, Microsoft and Amazon demonstrated that companies are continuing to spend aggressively on AI infrastructure, but investors have become far less willing to reward spending without evidence of improving returns.

That development raises the stakes considerably for SpaceX. Beyond financial performance, investors will likely focus on updates regarding semiconductor availability, AI compute capacity, regulatory approvals for Starship launches and the pace of commercial adoption across its AI businesses.

The company must also navigate growing regulatory scrutiny surrounding Grok after its image-editing capabilities generated controversy and prompted investigations and lawsuits in both the United States and Europe.

With its shares still trading at a premium valuation even after a steep correction, Tuesday’s earnings call is widely expected to serve as a referendum on whether management can convince investors that SpaceX’s massive investments in reusable rockets, AI infrastructure and satellite computing can ultimately generate the scale of earnings needed to support the world’s largest market capitalization.