Good point. Here is how @blevlabs prompts his AI. No tricks. Just talk to it like you talk to a human.
LLMS
-
Modern LLM Architecture: DeepSeek V3 vs Mistral 3
By
–
This blogpost is a masterpiece! The Big LLM Architecture Comparison From DeepSeek V3 to Mistral 3 Large: A Look At Modern LLM Architecture Design Link:
-
Gemini 3 Flash: Exploring the Latest Model Capabilities
By
–
This is all Gemini 3 Flash…will share the prompt shortly
-
Sora to use ChatGPT memories for personalized video generation
By
–
Sora will pull your memories from ChatGPT to generate this video! It would be cool if we could use this permanently, too!
— 🚨 AI News | TestingCatalog (@testingcatalog) 19 décembre 2025
"@soraofficialapp will use your ChatGPT memories to create a personalised holiday video for you." https://t.co/YmgzUUknEw pic.twitter.com/poKzacvDolSora will pull your memories from ChatGPT to generate this video! It would be cool if we could use this permanently, too! "
@soraofficialapp will use your ChatGPT memories to create a personalised holiday video for you." -
How AI Models Learn: Debunking Data Reproduction Myths
By
–
I’m tired of explaining how AI models learn. They don’t copy or reproduce data in the way you think they do.
-
Expectations for Gemini 3 Pro Release
By
–
This has nothing to do with 3.5 Gemini 3 Pro GA is expected to be another checkpoint of Gemini 3 Pro. In the same way as it was with Gemini 2.5
-

NotebookLM Updated with Gemini 3 Model
By
–

NotebookLM is now powered by Gemini 3 as well! This moment probably marks the time when the full lineup of Google AI products got upgraded. Gemini 3 everywhere
-
Pro Model Compares Favorably Against Flash Alternative
By
–
Pro is great but flash is not far behind.
-
Is Gemini 3 Pro the model behind this reasoning?
By
–
If reasoning – is it Gemini 3 Pro in this case?
-
Increasing Context Length and Model Capacity: Expensive Trade-offs
By
–
Either
a) you increase the context length and/or let the model crawl through stored data
b) you start of with a model that has large enough capacity and periodically update the weights
Both options are expensive, so I am not sure they actually want this right now.