91–100 of 157 posts
Setting up AWS so you can try it via Amazon Bedrock API is a hassle, so I made a step-by-step guide: https://ndurner.github.io/amazon-nova . It's 14+ steps!
Thanks for that. Are there any proxies that can communicate with bedrock and serve it via a OpenAI style api?
This is a digression, but I really wish Amazon would be more normal in their product descriptions. Amazon is rapidly developing its own jargon such that you need to understand how Amazon talks about things (and its existing product lineup) before you can understand half of what they're saying about a new thing. The way they describe their products seems almost designed to obfuscate what they really do. Every time the…
Once upon a time there were (and still are) mainframes (and SAP is similar in this respect). These insular systems came with their own tools, their own ecosystem, their own terminology, their own certifications, etc. And you could rent compute & co on them. If you think of clouds as being cross continent mainframes, a lot more things make a more sense.
> The Nova family of models were trained on Amazon’s custom Trainium1 (TRN1) chips,10 NVidia A100 (P4d instances), and H100 (P5 instances) accelerators. Working with AWS SageMaker, we stood up NVidia GPU and TRN1 clusters and ran parallel trainings to ensure model performance parity Does this mean they trained multiple copies of the models?
Part of it could also be that they'd prefer to move all operations to the in-house trn chips, but don't have full confidence in the hardware yet.
Def ambiguous though. In general reporting of infra characteristics for LLM training is left pretty vague in most reports I've seen.
More options/competition is good. When will we see it on https://lmarena.ai/ ?
When marketing talks about price delta and not quality of the output, it is DOA. For LLMs, quality is a more important metric and Nova would always try to play catch with the leaderboard forever.
Since Amazon are building their own frontier models, what's the point of their relationship with Anthropic ?
Different models have different strengths and weaknesses, especially here in the early days when models and their capabilities progress several times per year. The apps, programs, and systems based on models need to know how to exploit their specific strengths and weaknesses. So they are not infinitely interchangeable. Over time some of that differentiation will erode, but it will probably take years. AWS having cust…
I would add different errors as well. Here are two examples where GPT-4o and Claude 3.5 Sonnet cannot tell that "GitHub" is spelled like "GitHub".
GPT-4o: https://app.gitsense.com/?doc=6c9bada92&model=GPT-4o&samples...
Claude 3.5 Sonnet: https://app.gitsense.com/?doc=905f4a9af74c25f&model=Claude+3...
I don't think there will be one model that will rule them all, unless there is a breakthrough. If things continue on the same path, I think Amazon, Microsoft and Google will be the last ones standing, since they can provide models from all the major LLM players.
It's not clear what the use cases are for this, who is it aimed at.
Since Amazon are building their own frontier models, what's the point of their relationship with Anthropic ?
If you play all sides, you’ll always come on top.
Earlier quoted context omitted.
Thanks for that. Are there any proxies that can communicate with bedrock and serve it via a OpenAI style api?
You'd have to deploy it yourself, but there's this: https://github.com/aws-samples/bedrock-access-gateway
Since Amazon are building their own frontier models, what's the point of their relationship with Anthropic ?