I didn’t know mistral started serving deepseek-flash 4.1! This is awesome. A little more expensive than most providers though. $0.03/$0.007/$1.20 on mistral vs deepseeks own peak pricing of $0.03/$0.006/$1.20 (vs deepseeks off peak pricing of $0.015/$0.003/$0.60).
I prefer using a European provider though so this is actually cheaper than I can find elsewhere.
I feel exactly the same. Switchting to GLM for Le Chat without an option to prefer a Mixtral model leaves a bad taste. I hope this is temporary while the new model rolls out, but this is not filling me with confidence in either the model or the judegement of the company.
I’m not sure it speaks so poorly on the company. I’m not sure whether they’re using GLM 5.3 or 5.3 flash. The former is twice the price (probably better results?) and the latter is a quarter the price.
If they don’t end up switching models by the end of the year, I think that says a lot about the models. If they’re just waiting for an appropriately large or small model, or waiting to get large 4 put through its paces before switching, I think that’s a pretty ego-less way to go about it, and could speak pretty highly in a way.
Mistral large 3/medium 3.5/small 4 have all been pretty dated for a while so if they just needed a stopgap, I could understand. Now that large 4 is out though, we will see if that was the case.
Not sure it’s particularly odd. Several models of late have launched with an introductory promotional pricing, only to later lock in the promotional pricing.
Too many models now to remember specifics - I believe deepseek v4 did this and one of the gpt models (5? 5.1? 6?) did this if I recall correctly. I’m also fairly confident Google has had a few promotional periods on launch of the model, I seem to remember them starting the trend.
I’ve been using mistral-large-3 a tremendous amount. Price/speed/quality ratio was the best I could find. Been using the mistral-large-4 now since it came out for a few things, and while it’s not price competitive for the quality, it is really good. I’ve replaced a lot of my browser-based-chat-ui-llm-usage with mistral since it launched.
To me this implies they continued the training run for 4.1 and it improved. A 4.2 would imply they changed something about how it’s trained/how inference works, and a 5.0 would be a major update (who knows what would have changed).
I have been waiting for the next generation of mistral models for so long now. They were supposed to come out this summer… I hope the delay is because they’re making something better, not just because they’ve fallen too far behind.
Mistral models were the best for my use case when they came out, but they’re mostly almost a year old now. Hard to keep justifying using them, especially with all the price cuts this year on other models.
I’m building a language learning application [1]. It uses spaced repetition, built upon FSRS, an open source algorithm.
I’ve spent the last three years tweaking away at this algorithm. There are so many little details that have had to come together to make it an enjoyable experience. For example, take the one problem of balancing multiple language: What is the best way to balance multiple languages? How should you switch between them, and how often? What happens if you are over performing in one language and underperforming in another? What happens if you have more reviews in one language than another that isn’t in line with your priorities? How do you deal with competing learning speeds and language priorities? Spiky review burdens? What happens when you change your language priorities?
This is probably the most trivial part of the algorithm, and it took years of trial and error and dozens of iterations to get it to feel right. It requires knowing about SRS, the FSRS algorithm, language learning principles, and even then tons of experimentation, trial and error, and hundreds of days of daily usage.
I’ve asked LLMs every step of the way what to do. I’ve probably taken <1% of the advice I was given. Just asking an LLM “build an SRS app that balances multiple languages” would produce a result, but one that would definitely be terrible to use, and probably would end up slowing you down, not speeding you up. Explaining to it any of the issues that you would come across when using it will likely suggest to changes that will only make things worse, based on the thousands of terrible ideas I’ve seen confidently produced.
The hard part of a product is not the idea, but the experience of using it. The valueable work is normally not in the big feature specs, but the hundreds of little paper cuts that have all been smoothed over. The time saved is not because you don’t have to code it, but is from not having to become an expert in several different fields.
Phweww - 100% agree and am seeing that with my app as well! It's been one of the most joyful pieces of building it, is that I am really making something for myself and have been using it and continually adding the things that I want.
> I can't speak for other products, but on the type of features I work on, AI is blatantly incompetent
I do chat with the web interfaces of LLMs quite a bit as my workflow. I’m always baffled by how rarely it arrives at the correct solution.
I pose a question with code snippets, it gives me some incorrect answer. I pose a better question, and it gives me another wrong answer. Wash rinse repeat several times; not until the LLM gets the right answer, but until I pose the question specific enough that I can answer it myself.
I do find this loop of question refinement quite beneficial, but man it makes me terrified of vibe coded applications. 9/10 times the LLM doesn’t even get in the ballpark of the problem, let alone the solution.
Not only that, but even when you get the LLM to give you the right answer, if you ask it to turn that answer into actual code committed to your repo, it usually destroys something else in the process.
reply