Amazon Bedrock Is Now Generally Available
aws.amazon.com
Amazon Bedrock Is Now Generally Available
1–10 of 11 posts
Re: Amazon Bedrock Is Now Generally Available
#2Tangential, but I kinda miss the days when this would be SFM, Simple Foundation Models.
Re: Amazon Bedrock Is Now Generally Available
#3It says it has support for Llama 2, but in a deep page you can read that it is "coming soon".
Anyway, good to see support for serverless inference and painless train of Llama 2 models!
Re: Amazon Bedrock Is Now Generally Available
#4It says it has support for Llama 2, but in a deep page you can read that it is "coming soon". Anyway, good to see support for serverless inference and painless train of Llama 2 models!
Interesting timing given CloudFlare just released Workers AI, which actually has Llama 2 [1], not just coming soon. Great to see some competition.
Re: Amazon Bedrock Is Now Generally Available
#5Asking because I’m lazy: if I need to transcribe audio in real-time, is there a state of the art model I can plug into?
Re: Amazon Bedrock Is Now Generally Available
#6Asking because I’m lazy: if I need to transcribe audio in real-time, is there a state of the art model I can plug into?
https://github.com/ggerganov/whisper.cpp
https://github.com/Const-me/Whisper
I had fun with both of these. They will both do realtime transcription. Bit you will have to download the training data sets…
Re: Amazon Bedrock Is Now Generally Available
#7It says it has support for Llama 2, but in a deep page you can read that it is "coming soon". Anyway, good to see support for serverless inference and painless train of Llama 2 models!
Interesting timing given CloudFlare just released Workers AI, which actually has Llama 2 [1], not just coming soon. Great to see some competition.
Re: Amazon Bedrock Is Now Generally Available
#8Asking because I’m lazy: if I need to transcribe audio in real-time, is there a state of the art model I can plug into?
I saw whisper recommended, but I was curious how it compares to the other robust ASR systems (like nvidia's Nemo + Riva). I found this Twitter thread that seemed relevant: https://nitter.net/lunixbochs/status/1574848899897884672
Long story short, it depends on what you want to use it for. Different models and different training sets can help optimize for different things. Also, if you're in a domain with very uncommon speech patterns (think doctor shorthand or radio lingo), you'll need to understand how difficult it will be to customize generated models to do better in your space. I think Nemo + Riva does well at this; but I'm not as familiar with other options.
Re: Amazon Bedrock Is Now Generally Available
#9Asking because I’m lazy: if I need to transcribe audio in real-time, is there a state of the art model I can plug into?
I saw whisper recommended, but I was curious how it compares to the other robust ASR systems (like nvidia's Nemo + Riva). I found this Twitter thread that seemed relevant: https://nitter.net/lunixbochs/status/1574848899897884672 Long story short, it depends on what you want to use it for. Different models and different training sets can help optimize for different things. Also, if you're in a domain with very uncommo…
Yeah I fine tuned a model 2 years ago but it was a big pain and performance didn’t get better than 85%
Re: Amazon Bedrock Is Now Generally Available
#10If you want to use this but are tired of waiting for the SDKs to get updated you can do this:
```
curl -sS https://cdn.caylent.com/bedrock_new.zip > bedrock.zip
mkdir -p ~/.aws/models/
unzip bedrock.zip -d ~/.aws/models
```