3 ms·
I think the results might be underwhelming - AI providers need to turn a profit and can't subsidize, and they're working off of the commodity hardware everyone
by torginus 3mo ago
I think the results might be underwhelming - AI providers need to turn a profit and can't subsidize, and they're working off of the commodity hardware everyone does.
I wouldn't be surprised if they started offering potentiall bad quantizations with much reduced capability at lower prices (without telling the users, of course)
- user43928 3mo agoI would be surprised, considering that OpenRouter requires disclosing the quantization and shows automatic benchmarks to compare between providers for the same model.
- m00dy 3mo agoThe latter is a joke
- user43928 3mo agoI saw that it runs GPQA Diamond and TAU-Bench Airline and shows the results over a 32 day rolling average. Other than that they track Tool call error rate and Structured output error rate. I only discovered this today, and it seems like a good idea. What are the problems in practice?