Le Chat: Custom MCP Connectors, Memories

(mistral.ai)

397 points Anon84 | 2 comments | 04 Sep 25 11:04 UTC | HN request time: 0.481s | source

Show context

barrell ◴[04 Sep 25 11:36 UTC] No.45126116[source]▶

I recently upgraded a large portion of my pipeline from gpt-4.1-mini to gpt-5-mini. The performance was horrible - after some research I decided to move everything to mistral-medium-0525.

Same price, but dramatically better results, way more reliable, and 10x faster. The only downside is when it does fail, it seems to fail much harder. Where gpt-5-mini would disregard the formatting in the prompt 70% of the time, mistral-medium follows it 99% of the time, but the other 1% of the time inserts random characters (for whatever reason, normally backticks... which then causes it's own formatting issues).

Still, very happy with Mistral so far!

replies(11): >>45126199 #>>45126266 #>>45126479 #>>45126528 #>>45126707 #>>45126741 #>>45126840 #>>45127790 #>>45129028 #>>45130298 #>>45136002 #

1. thijsverreck ◴[04 Sep 25 11:46 UTC] No.45126199[source]▶

>>45126116 #

Any chance at fixing it with regex parsing or redoing inference when the results are below a certain treshold?

replies(1): >>45126444 #

2. barrell ◴[04 Sep 25 12:23 UTC] No.45126444[source]▶

>>45126199 (TP) #

It’s user facing, so will just have an option for users to regenerate the explanation. It happens rarely enough that it’s not a huge issue, and normally doesn’t effect content (I think once I saw it go a little wonky and end the sentence with a few random words). Just sometimes switches to mono space font in the middle of a paragraph, or it “spells” a word wrong (spell is in quotes because it will spell `chien` as `chi§en`).

It’s pretty rare though. Really solid model, just a few quirks

↑