Standing room only at the #H2O.ai booth at #ai4! Lots of excitement around our open source h2oGPT and LLM Studio demos. Stop by and see us! #h2oGPT #LLM #ai42023 #generativeAI
GENERATIVE AI
-
Fine-tuning SDXL models on Replicate platform
By
–
FYI https://
replicate.com/blog/fine-tune
-sdxl
… -
Fine-tuning Your Own Model Guide Released
By
–
Check out the blog post to learn more about fine-tuning your own model. Happy hacking!
-
Running Inference with Custom Concept Token in Prompts
By
–
Once it's done, you can run an inference on your model. Make sure to use the `TOK` token in your prompt (TOK is our default concept token). "Einstein having a pina colada on the beach in the style of TOK"
-
Getting Started: Collect 5-6 Images of Your Subject
By
–
To get started, collect your images. You'll want at least 5-6 images of your subject (for more info on sample data, check out our examples https://
github.com/replicate/cog-
sdxl/tree/main/example_datasets
…) -
Fine-tune SDXL with Barbie movie scenes and styles
By
–
You can fine-tune SDXL on styles and faces. Here's a model based on scenes from the Barbie movie: https://
replicate.com/fofr/sdxl-barb
ie
… -
Fine-tune SDXL open-source image model with your images
By
–
SDXL is the best open-source image model ever created. Now you can fine-tune it with your own images on Replicate.
-
Anthropic Joins Forbes Cloud100 List, CEO Discusses AI Progress
By
–
We are honored to be part of the annual
@Forbes #Cloud100 list! Our CEO Dario Amodei sat down with @kenrickcai to talk about how AI is remaking the cloud and notes that "understanding the pace of progress of technology" is key to disrupting the industry. -
Groq Achieves 100 Tokens Per Second Performance on Llama2
By
–
100 Tokens per second per user on #Llama2 from @MetaAI! This ultra-low latency performance could have a massive impact on workloads using #LLMs for everyone from artists to analysts, programmers to educators, all #GenAI and beyond. Book your demo to learn more: contact@groq.com pic.twitter.com/VWcSqPRm18
— Groq Inc (@GroqInc) 8 août 2023100 Tokens per second per user on #Llama2 from @MetaAI
! This ultra-low latency performance could have a massive impact on workloads using #LLMs for everyone from artists to analysts, programmers to educators, all #GenAI and beyond. Book your demo to learn more: contact@groq.com -
Groq Achieves 100 Tokens Per Second With Llama-2 70B
By
–
Announcement. @GroqInc is the first to accomplish 100 tokens per second, per user, running @MetaAI Llama-2 at 70B parameter size as an #LLM . No kernels or CUDA libraries necessary! Save thousands of developer hours with our deterministic Compiler methods.