Data and AI struggles

2,114 problems · 377 companies

LLM apps

Trouble with Mistral large

1 post from 1 person1d activeFading, +0 this weekunverified

Posts per day

Posts per day1 posts, Nov 5 to Nov 21

The posts behind it

GitHub issuesmistralai/mistral-vibe · Simon-LMNov 5

bug: Mistral Large 4 is offered `max` and not `medium`, but the API takes low, medium and high

**Component**: CLI **Summary** Expected: the thinking levels Vibe offers for Mistral Large 4 are the ones the API accepts: low, medium and high. Actual: the built-in model table gives Mistral Large 4 the levels of GLM 5.3, `("low", "high", "max")`...

1 reactions1 replies

Companies and products named

1 post
1 post
About this problem

Evidence

1 post from 1 person in 1 place. Mostly on mistralai/mistral-vibe. Tools named alongside: Reply.

Frustration Frustration 0 of 3· Seen on GitHub issues

How it was grouped

Posts about Mistral with no known issue, sharing the word "large". First post Nov 5, 2026, latest Nov 5, 2026. Unverified: fewer than 3 posts, or one person or place only. Method

Other Mistral problems

Mistral tables suffer from small files and skew

2 posts from 2 people on GitHub issues. Unverified.

Mistral users keep filing bugs

Top post: "Chromium and Firefox browser extension", one of 24 posts from 23 people on GitHub issues.

Trouble with Mistral vibe

Top post: "Vibe desktop on macOS and Windows", one of 11 posts from 10 people on GitHub issues.

Mistral upgrades break working setups

15 posts from 14 people on GitHub issues.

Mistral writes buggy code that is hard to review

12 posts from 10 people on GitHub issues.

Trouble with Mistral session

Top post: "/status session cost ignores prompt-cache discount on the Unified Harness (cached tokens billed at full input", one of 9 posts from 9 people on GitHub issues.

Mistral rate limits and usage caps block work

5 posts from 5 people on GitHub issues.

Mistral runs out of context on real work

5 posts from 5 people on GitHub issues.

Mistral fails to return valid structured output

3 posts from 3 people on GitHub issues.

Trouble with Mistral harness

Top post: "`enable_otel` exports no traces under the default harness, tracing is only wired into the legacy", one of 3 posts from 3 people on GitHub issues.

Trouble with Mistral return

Top post: "apply_chat_template ignores return_tensors=None for processors that default to tensors (e.g. Pixtral)", one of 3 posts from 3 people on GitHub issues.

Running Mistral locally hits memory and speed limits

2 posts from 2 people on GitHub issues. Unverified.

History

Momentum went from new to fading.

Nov 12, 2026

First seen with 1 post; the top post now: "Mistral Large 4 is offered `max` and not `medium`, but the API takes low, medium and high".

Nov 10, 2026