https://news.ycombinator.com/item?id=45200925

https://thinkingmachines.ai/blog/defeating-nondeterminism-in...

> As it turns out, our request’s output does depend on the parallel user requests. Not because we’re somehow leaking information across batches — instead, it’s because our forward pass lacks “batch invariance”, causing our request’s output to depend on the batch size of our forward pass.

tl;dr: the way inference is batched introduces non-determinism.

Most groups I work with stick to traditional automation/rules systems, but top-down mandates are pushing them toward frontier models for general tasks—which then get plugged into these workflows. A lot stays in sandbox, but you'd be surprised what's already live in fin services.

The authorities I cited (FSB/BIS/CFTC) literally just said last month AI monitoring is "still at early stage" cc https://www.fsb.org/2024/11/the-financial-stability-implicat...

Curious how you'd tackle that real-time changing reg?

That's not the way regulations work. Your compliance is measured against a fixed version of legislation.

Also the mistral medium model we tested had ~70% deterministic outputs across the 16 runs for the text to sql gen and summarization in json tasks- and it had reasoning on. Llama 3.3 70b started to degrade and doesn’t have reasoning. But it’s a relevant variable to consider