Photo: VCG
DeepSeek's V4 Flash model has officially moved from preview to general availability, sparking a wave of enthusiastic reactions across Chinese and overseas social media platforms which hailed its combination of near-frontier performance and ultra-low pricing, with many calling it a potential market disruptor.
Developers and AI enthusiasts have flooded the X platform with pointed comparisons, with some describing DeepSeek as offering Ferrari-level performance at bicycle prices and others hailing the open-source model for its accessibility and affordability.
X user @SLVR_CROW wrote: "Deepseek V4 Flash 0731 is why I will always love China. They the real OpenAI." Another user, @FoxRick01, noted: "Competition is healthy. Let's see what Anthropic answers after getting destroyed by Codex the whole week and now by Deepseek at 1/100th of the cost lol. They look now look like fiat salesmen charging ferrari prices."
Practical costs have fueled the hype. X user @tonbistudio reported: "Spent all day running DeepSeek-V4-Flash 0731 through multi-hour, challenging tasks and managed to spend only $0.31. Similar tasks last week with Kimi K3 and similar priced models cost me $35… (spoiler: the expensive models weren't 100x better)."
Xiaoyin Qu (@quxiaoyin), a Silicon Valley entrepreneur and founder of AI startup Tycoon, predicted broader market impact: "Prediction: Deepseek v4 flash will take over Claude as the no.1 winner in market share. It's the biggest story of 2026! Three reasons: 1. It's so cheap. 100x cheaper in unit token pricing. 2. It's so fast. Much much faster than Claude… 3. It's so cache-efficient… For something that's 500x cheaper AND 2-3x faster per task, why wouldn't Deepseek V4 Flash win."
Data from major platforms underpins the enthusiasm. On OpenRouter, the largest AI model aggregation platform, DeepSeek V4 Flash processed 7.1 trillion tokens in the most recent week (July 27-August 2), ranking first overall. Its sibling V4 Pro also held a top-five position.
Coding-focused platform OpenCode reported that DeepSeek V4 Flash handled 8 trillion tokens in a single day, with usage volume and new subscriptions both rising approximately 30 percent after the official launch.
Independent evaluation firm Artificial Analysis scored DeepSeek V4 Flash at 50 on its Intelligence Index—just one point behind GPT-5.6 Luna's 51. Yet the Chinese model's output pricing sits at $0.28 per million tokens, while even the recently discounted GPT-5.6 Luna remains at around $1.20.
A concept widely circulating summarized the competitive landscape: Codex and Claude set the upper limit, while DeepSeek sets the "kill line" — match it and there is no premium left; fall behind and you are out. In this framework, DeepSeek's latest V4-Flash version has simultaneously raised the bar for both capability and cost efficiency, fundamentally rewriting the rules of the game, Wang Peng, an associate research fellow at the Beijing Academy of Social Sciences, told the Global Times on Tuesday.
The "kill line" suggests that any model that is both weaker and more expensive than DeepSeek V4 Flash is considered commercially uncompetitive for many workloads.
Not all voices are purely celebratory. Some users caution that V4 Flash remains a smaller model and that real-world gaps versus the absolute frontier persist. Nonetheless, the dominant sentiment on X is one of excitement over a new era of AI economics in which price-performance, rather than pure capability, increasingly determines market share.
As Wang noted, the core of the AI race is no longer who has the most capable model, but who can empower real-world scenarios with AI.
Echoing that view, a Weibo user argued that AI may ultimately become less a standalone tool than a foundational layer of infrastructure. Like electricity, it could be embedded across factories, hospitals, vehicles and office systems, becoming a default capability underpinning every industry. Its biggest impact may not be giving people another chatbot, but providing the computing backbone that reshapes how society operates.
