Re:DeepSeek launching v4.1 flash cheaper and more capable th...
by packetlost on Wednesday September 09, 2026 @10:57AM
(#49627698)
> They’re nondeterministic at a fine level, but can be “deterministic” at a more general level: e.g. you might know that one model will always return properly formatted json when asked. That might not be true of the replacement, even if it is in general “better” and cheaper.
This isn't true. Even Sol messes up JSON formatting for me on occasion.
Do not delude yourself into thinking these things are reliable. They are not.
Re:DeepSeek launching v4.1 flash cheaper and more capable th...
by kamranjon on Wednesday September 09, 2026 @11:31AM
(#49628229)
Is nobody using structured outputs? They use constrained decoding at the generation stage to ensure the probability of tokens that would break the format are set to 0. I kinda figured everyone was doing this at this point.
Re:DeepSeek launching v4.1 flash cheaper and more capable th...
by nolok on Wednesday September 09, 2026 @11:53AM
(#49628577)
If you use a large enough volume, you will know this isn't fully reliable. You might get json, and it might not match what the model actually sent because the last layer cut it up to match what you want. At the end, not json, or json but not really matching what the model wanted, it's sort of the same issue: when you use them you HAVE to assume they can have a brain fart. That's fine, just code around it.
Re:DeepSeek launching v4.1 flash cheaper and more capable th...
by wongarsu on Wednesday September 09, 2026 @11:57AM
(#49628636)
Most serious providers are now supporting structured outputs in a reasonable way for all model configs. But for example on ollama structured outputs are still incompatible with tool calling and with reasoning
Re:DeepSeek launching v4.1 flash cheaper and more capable th...
by Zopieux on Wednesday September 09, 2026 @03:15PM
(#49632507)
Friends Don't Let Friends Use Ollama. https://sleepingrobots.com/dreams/stop-using-ollama/
Re:DeepSeek launching v4.1 flash cheaper and more capable th...
by gcanyon on Wednesday September 09, 2026 @12:49PM
(#49629472)
We use structured output and To my knowledge it has never failed (millions of data points). There seem to be two classes of people: those doing productive work with LLMs, and those who only get replies insulting their mothers…
Just the risk of such a thing means regression testing every time you update the model, and you want to be able to run that testing on your schedule rather than having it forced on you.