Skip to content
Local AI Wizard
Your machine
Theme

Compare local AI models

Up to three models side by side, with sources.

Phi-4Mistral Small 3.2 24B

2 of 3 models

Memory needed

Q4 · each model's own figure, before any machine's reserve

No score in LiveBench 2026-06-25 for these models (not drawn, not zero) Not in LiveBench 2026-06-25

Q4 planning estimate: weights plus working room, before any machine's reserve; not a local measurement.

Reported by each publisher

Same benchmark name on each model card, values as printed. Each publisher chose its own prompts, shots and harness, so these are not comparable and never ranked.

Results each publisher reports under the same benchmark name
BenchmarkPhi-4Mistral Small 3.2 24B
MMLU84.8 (simple-evals)80.50%

Specifications side by side

10 of 10 specifications

Specifications side by side
SpecificationPhi-4Mistral Small 3.2 24B
PublisherMicrosoftMistral
LicenseMITApache 2.0
Task and abilitiesText & chatText & chatVision
Parameters14 B24 B
Memory needed · Q4 · model onlyNeeds ~11 GBNeeds ~19 GB
Disk · Q4 weights~9 GB~15 GB
Context · tokens16,384131,072
LiveBench rank · open modelsNo score in this LiveBench releaseNo score in this LiveBench release
Measured speedNo published tokens-per-second measurement for this checkpoint7 published rows
SourceModel cardModel card

See which of these models run on your machine, well or slowly.

Published comparisons

Written comparisons with the same evidence, sources and limits.