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

Mistral Small 4 (119B-2603) vs Qwen3 30B-A3B Instruct 2507

Mistral Small 4 (119B-2603) scores higher on the intelligence index — 11 against 10. Qwen3 30B-A3B Instruct 2507 is the smaller download at 19 GB, so it runs on cheaper hardware.

Mistral Small 4 (119B-2603)Qwen3 30B-A3B Instruct 2507
Intelligence index1110
ClassBelow every hosted tierBelow every hosted tier
Weights74 GB19 GB
QuantisationQ4_K_MQ4_K_M
Parameters119.4B (6B active)30.5B (3.3B active)
Max context1024k256k
API price per 1M$0.15 in / $0.6 out$0.048 in / $0.193 out
LicenceApache 2.0Apache 2.0
Cheapest machine that runs itStrix Halo Framework Desktop, 128GB $3,449Strix Halo Framework Desktop, 32GB $1,269
Summarising not rated good
Translation not rated good
Everyday coding not rated good
Reasoning & maths not rated usable
Agentic work not rated usable

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.