Live data from Hacker News

Amazon Nova

aws.amazon.com

121–130 of 157 posts

Re: Amazon Nova

#121
post #110

Earlier quoted context omitted.

It’s fascinating that Amazon is investing heavily in Anthropic while simultaneously competing with them.

It’s fascinating that Amazon Web Services have so many overlapping and competing services to achieve the same objective. Efficiency/small footprint was never their approach :D For example, look how many different types of database they offer (many achieve the same objective but different instantiation) https://aws.amazon.com/products/?aws-products-all.sort-by=it...

To quote, “right tool for right job”.

Re: Amazon Nova

#122

A rough idea of the price differences... Per 1k tokens Input | Output Amazon Nova Micro: $0.000035 | $0.00014 Amazon Nova Lite: $0.00006 | $0.00024 Amazon Nova Pro: $0.0008 | $0.0032 Claude 3.5 Sonnet: $0.003 | $0.015 Claude 3.5 Haiku: $0.0008 | $0.0004 Claude 3 Opus: $0.015 | $0.075 Source: AWS Bedrock Pricing https://aws.amazon.com/bedrock/pricing/

does anyone know performance benchmark

Re: Amazon Nova

#123
post #59

A rough idea of the price differences... Per 1k tokens Input | Output Amazon Nova Micro: $0.000035 | $0.00014 Amazon Nova Lite: $0.00006 | $0.00024 Amazon Nova Pro: $0.0008 | $0.0032 Claude 3.5 Sonnet: $0.003 | $0.015 Claude 3.5 Haiku: $0.0008 | $0.0004 Claude 3 Opus: $0.015 | $0.075 Source: AWS Bedrock Pricing https://aws.amazon.com/bedrock/pricing/

I suggest you give the price per million token as seems to be the standard.

From my personal table https://i.imgur.com/WwL9XkG.png

Price is pretty good. I'm assuming 3.72 chars/tok on average though.. couldn't find that # anywhere.

Re: Amazon Nova

#125
post #74

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!

This appears to be a way to steal and harvest aws credentials. No one should be following any of these steps.

Do you have any evidence for this accusation?

This is a guide for the casual observer who wants to try things out, given that getting started with other AI platforms is so much more straightforward. It's all open source, with transparent hosting, catering to any remaining concerns someone interested in exactly that may have.

Re: Amazon Nova

#126
post #74

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!

Nice! FWIW, The only nova model I see on the HuggingFace user space page is us.amazon.nova-pro-v1:0. I cloned the repo and added the other nova options in my clone, but you might want to add them to yours. (I would do a PR, but... I'm lazy and it's a trivial PR :-)).

OK! I only add what people are interested in, so noted with thanks - will do! :-)

Re: Amazon Nova

#127
post #116
post #74

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!

I'm so confused on the value prop of Bedrock. It's seems like it wants to be guardrails for implementing RAG with popular models but it's not the least but intuitive. Is it actually better than setting up a custom pipeline?

The value I get is: 1) one platform, largely one API, several models, 2) includes Claude 3.5 "unlimited" pay-as-you-go, 3) part of our corporate infra (SSO, billing, ... corporate discussions are easier to have)

I'm using none to very little of the functionality they have added recently: not interested in RAG, not interested in Guardrails. Just Claude access, basically.

Re: Amazon Nova

#130
post #125

Earlier quoted context omitted.

This appears to be a way to steal and harvest aws credentials. No one should be following any of these steps.

Do you have any evidence for this accusation? This is a guide for the casual observer who wants to try things out, given that getting started with other AI platforms is so much more straightforward. It's all open source, with transparent hosting, catering to any remaining concerns someone interested in exactly that may have.

The most common way for an AWS account to be hacked, by far, is mishandling of AWS IAM user credentials. AWS has even gone so far as to provide multiple warnings in the AWS console that you should never create long-lived IAM user credentials unless you really need to do so and really know what you are doing (aka not a “casual observer who wants to try things out”).

This blog post encourages you to do this known dangerous thing, instructs you to bypass these warnings, and then paste these credentials into an untrusted app that is made up of 1000+ lines of code. Yes, the 1000+ lines of code are available for a security audit, but let’s be real: the “casual observer who wants to try things out” is not going to actually review all (if any) of the code, and likely not even realize they should review it.

I give kudos to you for wanting to be helpful, but the instructions in this blog (“do this dangerous thing, but trust me it’s okay, and then do this other dangerous thing, but trust me it’s okay”) is exactly what nefarious actors would ask of unsuspecting victims, too, and following such blog posts is a practice that should not be generally encouraged.

Post reply on HN