The best AI companies are going to come from NYC — not SF. (with a few notable exceptions)
@mattshumer_
-
LLaMA 3 Release Expected to Surpass Mistral Models
By
–
LLaMA 3's will start to drop next week. Assuming there's a 7B version, I'm expecting it to far surpass the current Mistral model.
-
Sweeps Training System for Axolotl Open Source
By
–
Built a simple sweeps system for training with Axolotl… would anyone want this/find this useful? cc @winglian
-
Skipping College for Early Tech Career Advantage
By
–
Hahah yeah, but I skipped most of college, so I had a head start
-
OpenRouter Implementation of AI-Oracle Tool Released
By
–
Thanks to @abdiisan for implementing an OpenRouter version of `ai-oracle`! https://
x.com/abdiisan/statu
/abdiisan/status/1777423304086753725
… -
Combining Multiple AI Models for Improved Results
By
–
It's not perfect, but on average, it should improve results significantly compared to using models individually. If anyone wants to improve it, there a lot of gains to be made by adding context about the strengths/weaknesses of each model in the final prompt.
-
AI-Oracle: Open-source tool combining Claude, GPT-4, Perplexity
By
–
Open-sourcing `AI-Oracle`.
— Matt Shumer (@mattshumer_) 8 avril 2024
Generates better responses than Claude 3 Opus.
A very simple approach that combines the abilities of Claude 3, GPT-4, and Perplexity to provide better results than any could provide on their own.
Seriously — it's dumb simple.
Notebook in thread: pic.twitter.com/8HQBUEiHoIOpen-sourcing `AI-Oracle`. Generates better responses than Claude 3 Opus. A very simple approach that combines the abilities of Claude 3, GPT-4, and Perplexity to provide better results than any could provide on their own. Seriously — it's dumb simple. Notebook in thread:
-
Multi-Model AI Orchestration Combining Claude GPT-4 Perplexity
By
–
How does it work? The process is super simple. We simply query each model individually:
– Claude 3 Opus for reasoning + personality
– GPT-4 for reasoning
– PPLX for freshness/up-to-date info Then, Claude combines the strengths of each and responds with a final, ideal output. -
Fine-tuning Models: When to Use It Versus Prompting
By
–
If you haven’t hit PMF, you should not be fine-tuning models. Just prompt. Only exceptions:
– Speed is required for your use-case (and you need a model more powerful than Mixtral)
– Opus/GPT-4 cannot do your task or are too expensive per use -
Privacy Concerns Don’t Drive Youth Toward Self-Hosted Solutions
By
–
And for the people who keep saying that privacy will drive people towards self-hosted solutions: I’m 24. People my age have grown up with tech, and are generally more capable at using it than older generations. The amount of non-tech friends that I’ve seen upload pictures of