
Midjourney V7
Rebuilt-from-scratch image model with Draft Mode, Omni Reference, and sharper prompt adherence.
About
Launched in alpha on April 3, 2025 and default from June 17, V7 is Midjourney's first full architecture rebuild since V6 — not an iteration. It reads prompts more literally, renders hands, anatomy, and fine detail far more coherently, and keeps the signature cinematic look. New capabilities include Draft Mode (roughly 10x faster at a fraction of the cost for ideation), Omni Reference (--oref) generalizing character reference to any subject, and personalization on by default. It also added image-to-video clips up to 21 seconds.\n\nThe honest caveats: still Discord-first (a web app exists but the culture is Discord commands), closed weights, no API for most users, and $10/mo entry. Exact typography and rigid layouts remain weak spots — Ideogram or GPT Image win there. Some edit actions still fall back to V6.1 surfaces, and V8 entered alpha in March 2026, so V7's reign as default is already being succeeded.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 9/10
Scored 9/10 (verified tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 51; Trending score 89/100 from researcher assessment.