Sunk Cost sunkcost.ai Data checked 2026-09-03

Llama 4 Scout 17B-16E vs Ministral 3 14B

Neither has a clear lead on the index. Ministral 3 14B is the smaller download at 8.2 GB, so it runs on cheaper hardware.

Llama 4 Scout 17B-16EMinistral 3 14B
Intelligence index66
ClassBelow every hosted tierBelow every hosted tier
Weights65 GB8.2 GB
QuantisationQ4_K_MQ4_K_M
Parameters108.6B (17B active)14B
Max context10240k256k
API price per 1M$0.1 in / $0.3 out$0.2 in / $0.2 out
LicenceLlama 4 Community LicenseApache 2.0
Cheapest machine that runs itStrix Halo Framework Desktop, 128GB $3,449Mac mini M6, 24GB $1,099
Summarising good not rated
Translation good not rated
Everyday coding usable not rated
Reasoning & maths usable not rated
Agentic work usable not rated

Ratings are coarse on purpose. Speeds and pay-back depend on the machine — open either model's page for the full list, or see both against the frontier. Context is 32k throughout.