Deepseek V3.2 Exp: CHEAPEST Model Ever That Is Quite POWERFUL! Best Opensource Model?
Updated: October 9, 2025
Summary
The video covers updates in the AI space including Gemini 3.0 leaks and GLM 4.6 release, as well as the introduction of Deep Seek V3.2 with API cost reduction. It discusses the trade-offs in performance and model generation of Deep Seek V3.2, emphasizing the importance of experimental models. Comparison is made between Deepseek V3.1 Terminus model and Gemini 2.5 Pro, showcasing cost-effectiveness and performance. Viewers are encouraged to subscribe to the World of AI newsletter for weekly updates on AI developments and pricing structures, with details on the pricing structure of AI models and accessing them through API platforms provided in the video. AI model output quality is demonstrated through code generation and applications, highlighting low costs and high efficiency in achieving desired outputs. The video also showcases AI reasoning and problem-solving abilities through sequential problem-solving examples and evaluation of logic in solving puzzles.
AI Space Updates
Highlights of the AI space updates including Gemini 3.0 leaks, GLM 4.6 release, Deepseek V3, and cost reductions in API usage.
Deep Seek V3.2 Release
Introduction of Deep Seek V3.2 with API cost reduction and focus on important elements for cheaper and enhanced quality usage.
Deep Seek V3.2 Trade-offs
Discussion on the trade-offs of Deep Seek V3.2 in terms of performance and model generation, emphasizing the need for experimental models.
Benchmark Comparison
Comparison of Deepseek V3.1 Terminus model with benchmarks, highlighting its cost-effectiveness and performance against Gemini 2.5 Pro.
World of AI Newsletter
Encouragement to subscribe to the World of AI newsletter for weekly updates on AI developments and pricing structures of AI models.
Model Pricing Structure
Detailed overview of the pricing structure for AI models, including input and output token costs and the affordability of proprietary models.
API Access and Credits
Information on accessing AI models through API platforms, different quantiz models, Kilo code usage, and $25 of free credits offered.
Model Output Quality
Explanation of model output quality and its comparison to other models, demonstrating its performance in generating code and applications.
Generating Applications
Process of generating applications using AI models, testing the SAS landing page, and achieving desired outputs with low costs and high efficiency.
AI Logic and Problem Solving
Illustration of AI reasoning and problem-solving abilities using sequential problem-solving examples and evaluation of logic in solving puzzles.
FAQ
Q: What are the highlights in the AI space updates mentioned in the file?
A: The highlights include Gemini 3.0 leaks, GLM 4.6 release, Deepseek V3, and cost reductions in API usage.
Q: What is the new version introduced in the file and what key features does it offer?
A: The new version introduced is Deep Seek V3.2, which focuses on important elements for cheaper and enhanced quality usage with API cost reduction.
Q: What is the main discussion point regarding Deep Seek V3.2?
A: The discussion revolves around the trade-offs in terms of performance and model generation, emphasizing the need for experimental models.
Q: How is Deepseek V3.1 Terminus model compared in the file?
A: The comparison highlights its cost-effectiveness and performance against Gemini 2.5 Pro.
Q: What is being encouraged in terms of AI updates?
A: Subscription to the World of AI newsletter for weekly updates on AI developments and pricing structures of AI models is being encouraged.
Q: What details are provided about the pricing structure for AI models in the file?
A: Information about input and output token costs, affordability of proprietary models, and accessing AI models through API platforms, quantiz models, Kilo code usage, and free credits offered is provided.
Q: How is the model output quality discussed in the content?
A: The content explains the performance in generating code and applications and compares it to other models.
Q: What process is illustrated for generating applications using AI models?
A: The file details the process of testing the SAS landing page and achieving desired outputs with low costs and high efficiency.
Q: How are AI reasoning and problem-solving abilities portrayed in the file?
A: They are illustrated through sequential problem-solving examples and evaluation of logic in solving puzzles.
Get your own AI Agent Today
Thousands of businesses worldwide are using Chaindesk Generative
AI platform.
Don't get left behind - start building your
own custom AI chatbot now!