mistral-large-3 vs mistral-medium-3-5

Capability for the money

Not the same tier: mistral-medium-3-5 leads mistral-large-3 by 69.6 points on tool use (τ²-bench), for 4.0× the price.

1. Following instructions

   extraction, classification, sticking to a format

   IFBench
   mistral-large-336.2%

   mistral-medium-3-568.8%

   mistral-medium-3-5 +32.6

2. Tool use

   multi-step agent work against real tools

   τ²-bench
   mistral-large-324.6%

   mistral-medium-3-594.2%

   mistral-medium-3-5 +69.6

3. Hard reasoning

   graduate-level science and multi-step analysis

   GPQA Diamond
   mistral-large-368.0%

   mistral-medium-3-574.8%

   mistral-medium-3-5 +6.8

Averaged over the benchmarks both models sat, mistral-large-3 returns 57.2 points of pass rate per dollar of blended price against 26.4 — the ratio that decides whether the dearer model earns its rate.

Each figure is a pass rate: the share of that test set the model answered correctly, measured by a third party rather than self-reported by the lab. Rows are different tests of different difficulty, so compare a model against the other model, not one row against the next. No published coding score for one of these models, so that lens is omitted. And a benchmark is not your workload — it narrows the shortlist, it does not pick for you.