Alt history: What if OpenAI & Anthropic open-sourced GPT-4 & Claude 3.5 (including data & algorithms)? Would it:
Accelerate AGI?
Boost AI adoption?
Spur new distilled models?
Enhance/decrease AI safety?
Impact revenue (+/-)?
OPEN SOURCE
-
Open-sourcing GPT-4 and Claude 3.5: AGI acceleration and safety implications
By
–
-
US Tech Dominance and Open Source Rise: A Parallel Evolution
By
–
The period during which US tech became most dominant is the period during which open source became dominant. These things are obviously not inconsistent.
-
Try Out MiniCPM-V Gradio Demo on GitHub
By
–
try out the @gradio demo: https://t.co/9C6rsofy7B https://t.co/8SNzHdNK8A pic.twitter.com/7sj9d5xTdw
— AK (@_akhaliq) 7 août 2024try out the @gradio demo: https://
github.com/OpenBMB/MiniCP
M-V?tab=readme-ov-file#chat-with-our-demo-on-gradio-
… -
MiniCPM-V-2_6 Model Space on Hugging Face
By
–
https://
huggingface.co/spaces/openbmb
/MiniCPM-V-2_6
… -
Sharing Scientific Links in 2024
By
–
As promised: https://nature.com/articles/s41586-024-07566-y
… https://arxiv.org/html/2311.16822v2
… -
Dangers of LLM-Generated Training Data
By
–
One of the worst mistakes many AI teams are making today: Using data generated by LLMs to train machine learning models or create a 'self-learning' training loop. While the idea may seem appealing and
-
fh-bootstrap framework seeks pull requests for FastHTML implementation
By
–
A PR to this would be great. fh-bootstrap is what http://
about.fastht.ml is implemented with. -
LLaVA-OneVision: Open Large Multimodal Models for Visual Task Transfer
By
–
LLaVA-OneVision Easy Visual Task Transfer discuss: https://
huggingface.co/papers/2408.03
326
… We present LLaVA-OneVision, a family of open large multimodal models (LMMs) developed by consolidating our insights into data, models, and visual representations in the LLaVA-NeXT blog series. Our -
Building a Bignum Library for Mathematical Computing
By
–
Making a bignum library for fun https://
bit.ly/3xRELub
#AI #MachineLearning #DeepLearning #LLMs #DataScience