
Midjourney's 2026 in full: V8 alpha to the V8.2 default, native 2K without upscaling, 50% faster and 25% cheaper renders, and 21-second video.
V8.2 became the default on July 24 β an aesthetics-and-personalization release: bolder, more sophisticated output, and personalization that reads your taste far better. The year's workhorse update was V8.1 (April 30): native HD 2K images with no separate upscale step (HD renders 3Γ faster), standard generations 50% faster and 25% cheaper, and markedly better text rendering. Video generation now runs 5 to 21 seconds per clip with improved motion coherence.
| Date | Release | Highlights |
|---|---|---|
| FebβMar 2026 | V8 alpha | New personalization system, new edit model |
| Apr 30, 2026 | V8.1 | Native 2K (no upscale step) Β· HD 3Γ faster Β· standard gens +50% speed / β25% cost Β· better text Β· Image Prompts restored Β· Prompt Shortener Β· updated Describe |
| Jul 24, 2026 | V8.2 (default) | Aesthetics, image quality and personalization β "more creative, bold, sophisticated" output |
V8 Alpha (March 17, 2026) ran on a completely rewritten codebase, not an incremental retrain β which is why the gains are structural: 5Γ faster generation, native 2K HD, and dramatically improved text rendering. Backward compatibility held: existing V7 styles and moodboards carried over intact.
The economics deserve emphasis: V8.1 simultaneously cut cost and raised speed and baseline resolution β the rare update that improves every axis at once. If your workflows still route through a separate upscale pass, they're a version behind.
One counterweight to the "everything got cheaper" story: on V8, --sref, moodboards, --hd and --q 4 run 4Γ slower and consume 4Γ the GPU time of a standard job. Users who habitually run style references saw their effective allowance shrink even as headline prices fell. Explore with plain prompts; apply references once composition locks.
V8's arc is less about raw quality (already the artistic benchmark) and more about your aesthetic: the personalization system introduced in the V8 alpha and refined through 8.2 learns taste from your ratings and history. In practice, two users typing the same prompt now get meaningfully different β and individually better β results. That's a moat competitors can't copy with a bigger model.
Personalization is the one advantage a competitor can't close by scaling β accumulated, per-user, non-transferable. A larger model elsewhere still doesn't know what you like.
Midjourney's video mode animates generated or uploaded images β clips start at 5 seconds, extendable to 21 seconds, with V8.1 improving motion coherence. It's not competing with Veo 3.1 or Runway Gen-4.5 for cinematic work; it's the fastest path from a Midjourney still to a living visual, which for social and concept work is often exactly enough. Our archived V7 video-model review tracks where this line started.
The constraint that defines the mode: slow beats fast. Push-ins, gentle parallax, drifting particles and fabric hold coherence; limb articulation, whip pans and multiple moving subjects still break. Under-drive the motion and the usable-clip rate roughly doubles.
Midjourney video doesn't appear in Artificial Analysis' top 10 β currently led by ByteDance Seedance 2.0, Alibaba HappyHorse-1.0 and Veo 3.1 β because it isn't playing that game. Those models compete on fidelity, length, resolution and price per second.
A look people already chose. If your visual identity is Midjourney-built, no ranking changes the fact that only Midjourney animates your images in your style. That's narrower and more defensible than a benchmark position.
Subscription tiers span $10β$120/month across four levels. V8.1's 25% cheaper standard generations effectively raised every plan's included volume β but budget as standard jobs Γ 1, reference/HD jobs Γ 4, because the 4Γ multiplier is what actually determines which tier you need.
V8.1's rendering improvement finally makes short labels and posters viable without post-editing. Paragraphs still fail; treat the capability as "signage," not "typography."
Prompts largely carry over. The V7 tutorial remains as an archived reference with parameter differences noted, and the current complete guide covers V8.2 workflows end to end.
Stop. Native 2K since V8.1 means the separate upscale pass is pure waste β a step that costs time and adds nothing. Workflows built before April 30 almost certainly still contain it.
HD renders 3Γ faster than the V7-era baseline. Schedules built on old render times are over-provisioned, and the freed time is better spent on curation than on waiting.
Short labels and posters now render acceptably in-model. The round-trip to a design tool is still required for real typography, but not for a two-word overlay.
This is the one that got worse. The 4Γ GPU multiplier on --sref, moodboards, --hd and --q 4 means habitual reference use now consumes allowance far faster than the headline price cuts suggest. Restructure to explore plain and finish referenced.
Sources: Midjourney official version docs Β· TechJack V8.1 guide Β· ImageToolsAI news tracker Β· 9to6AI V8.1 review
Last updated: July 29, 2026
Anthropic's 2026 in business terms: the $65B Series H at a $965B valuation, confidential IPO filing, $47B run-rate, and enterprise adoption of Claude.
Every major ChatGPT update through July 2026: the GPT-5.5/5.6 transition, Dreaming V3 memory, Lockdown Mode, and what free vs paid gets you now.
Every major Claude update through July 2026: the Claude 5 family, Reflect dashboard, MCP upgrade, Claude Code milestones, pricing and access.
Copilot's 2026: Agent Mode general availability, agentic code review that ships fix PRs, the Copilot CLI, desktop app, and much faster indexing.
Every major Gemini update through July 2026: the 3.5/3.6 Flash generation, Gemini Omni multimodal creation, Nano Banana 2, and the agentic-era strategy.
Notion's 2026: autonomous Custom Agents, the credit billing switch on May 4, agents that read your files, 35β50% cheaper runs, and what it costs now.
OpenAI's 2026 from the platform side: the GPT-5.x release-and-retire cadence, the Sora shutdown, API changes, and where the company stands mid-year.
Perplexity's browser year: Comet free on every platform since March, iPad and iOS upgrades, Deep Research generating decks, and enterprise controls.