DeepSeek V4 incoming: cheap Chinese rival to Kimi K3 and Fable
Early videos show V4 matching Kimi K3 and Fable at a fraction of the cost.
DeepSeek is about to drop its fourth-generation model, V4, and early indicators point to a seismic shift in the AI model landscape. According to a Reddit post that has since gone viral, leaked videos show V4 performing at a level comparable to Kimi K3 and Fable — two highly regarded models — but at a drastically lower price point. The poster noted that Fable in particular may become obsolete once V4 launches, suggesting that DeepSeek is positioning V4 as a cost-effective replacement for premium AI models.
While DeepSeek has not confirmed an official release date, the community is buzzing with speculation. The claim that V4 will be “cheap” aligns with DeepSeek’s previous strategy of offering high-performing models at low cost. If V4 matches the quality of Kimi K3 and Fable, it could democratise access to advanced AI for startups and developers who previously had to pay premium prices. However, until a concrete benchmark or pricing announcement appears, these claims remain unverified. The AI world is watching closely.
- DeepSeek V4 is reportedly approaching release with leaked videos showing competitive performance.
- Performance is said to be on par with Kimi K3 and Fable, two established AI models.
- The model is expected to be significantly cheaper than current rivals, potentially replacing Fable.
Why It Matters
If true, DeepSeek V4 could slash AI costs and upend the market dominated by premium models.