I used Cmd-R on large documents, and it performed better than GPT4 for me! Yeah, NIAH correlates with that. You know how you can mention a custom GPT today? I would love to see the ability to make multiple mentions to intelligently direct flows based on each GPT's abilities.
@skirano
-
Future AI Capabilities: RAG, Context, Agents and Prompt Optimization
By
–
– Better RAG on large documents (the new @Cohere model is incredible for that)
– Larger token context
– Some sort of agentic behavior
– Ability to improve its own prompts
– Orchestrate subagents intelligently
– Multiple custom GPT tags -
Best Time Ever for Software Engineers and Problem Solvers
By
–
I actually disagree. In my opinion, there has never been a better time to be a software engineer or someone who is a true problem solver via code.
-
Diffusion-based generative model trained on product images
By
–
No, this is completely generated from scratch with a diffusion-based technique with a model trained on the product!
-
EverArt AI Achieves Record Single-Day Revenue Milestone
By
–
Yesterday was our biggest single-day revenue at @everartai so far. Thanks for believing in us and allowing us build these tools for you, we can’t wait to share what’s next!
-
Ideas Over Code Skills in the Age of AI
By
–
In an era when coding skills and speed are just a commodity, your ideas are more valuable than ever.
-
Claude 3 and Gemini 1.5 Pro Release Sparks Shared Ideas
By
–
This is awesome, Andrew! I feel like we all had the same idea after Claude 3 and Gemini 1.5 Pro dropped lol
-
Stable Diffusion MLX Repository Enhanced for SD Cascade
By
–
Edit the Stable Diffusion MLX repository, so that it can support SD Cascade for image generation!
-
Scaling Learning Rate with Batch Size in Model Training
By
–
I think it really depends on the batch size you are using. Let's say you are multiplying the batch size by x, then you should multiply the learning rate by sqrt(x)
-
Different Approaches to Same AI Problems in Open Source
By
–
Lol, classic! Just checked his repo, it seems to be quite a different approach. Glad we can get this level of diversity of thoughts around the same problems.