One thing that these nerfing conspiracy theorists fail to realize is that Anthropic isn’t the only organization that directly serves Claude inference.
My company uses Claude models exclusively via Azure and AWS bedrock, which have their own licensed copies of the weights.
If all these people are so convinced Anthropic is nerfing models, have they tried other inference providers? Do they think the nerfing is coordinated across independent providers? Why wouldn’t any of these nerf-benches use these comparison points or even talk about them?
I think the most likely conclusion by far is that this is a psychological phenomenon.
This is a good point but is it actually the case that inference providers directly control the inference code and weights, as opposed to being a hardware/infra provider? It seems like it would require a nontrivial amount of work, I doubt the inference is in any kind of standardized form. Moreover it seems that this would risk things like degradation if the model provider used different inference techniques (e.g. 16-bit vs 32-bit floats)
semiquaver · · focus · HN ↗
My company uses Claude models exclusively via Azure and AWS bedrock, which have their own licensed copies of the weights.
If all these people are so convinced Anthropic is nerfing models, have they tried other inference providers? Do they think the nerfing is coordinated across independent providers? Why wouldn’t any of these nerf-benches use these comparison points or even talk about them?
I think the most likely conclusion by far is that this is a psychological phenomenon.
dooglius · · focus · HN ↗