17 categories · 102 tools

Every job, ranked.

Pick a job. Each page ranks every tool we track for it, with price sources, local vs cloud and platforms.

Speech-to-text & dictation

08

Text-to-speech & voice

05

Image generation

08

Video generation & editing

09

Music

03

Coding assistants

09

Chatbots & assistants

05

App & site builders

08

Writing

09

Meeting notes

06

Research & search

09

Presentations

05

Design & logos

08

Automation & agents

05

Productivity & ADHD

08

File conversion & PDFs

06

Game making

06
How we rank

Scores you can check.

  • Quality (1–5). We start from a hands-on editorial score, then move it with public benchmarks when the tool's own model is ranked: LMArena (text, agents, coding, webdev, image and video arenas), Artificial Analysis (Intelligence Index, speech, music and video arenas) and SWE-bench Verified. #1 on a board adds 0.3, top 3 adds 0.2, top 10 adds 0.1, #11–20 holds, #21–40 takes 0.2 off and lower takes 0.4 off. We average across boards.
  • User ratings. Product Hunt reviews (10+ reviews) and US App Store ratings (50+ ratings), and Trustpilot scores (50+ reviews) that we read from the live page, not from search previews. An average of 4.8+ adds 0.2 to ease, 4.6+ adds 0.1, under 4.3 takes 0.2 off and under 4.0 takes 0.4 off. When no benchmark exists, half of that also moves quality. GitHub stars and Hugging Face downloads are listed as adoption signals but don't change scores.
  • Value (1–5) is cost per quality point: the lowest paid monthly price on the official pricing page divided by quality. $1.50 or less scores 4.5, $3 or less 4, $5 or less 3.5, $8 or less 3, $15 or less 2.5, more than that 2. A usable free tier adds 0.5, and free or open-source tools score 5. One-time licences are spread over 24 months. If we couldn't read a price, value falls to a neutral 2.5 (3.5 with a free tier) and the tool is marked unverified.
  • Confirmed removals. If a reliable source confirms that a vendor removed a core feature (for example, Udio turning off downloads), ease drops 0.5 and value drops 1. The source is listed on the tool's page.
  • Best overall is mostly quality. Best value is 60% value and 40% quality.
  • What's not in yet. G2, Capterra and Reddit block automated reading, and we haven't scored X sentiment, so none of those affect scores. We'll add them only when we can cite them.
  • Right now: 172 cited data points across 102 tools. Plan prices were checked 2026-10-14: 79 fully verified, 16 partly verified, 7 unverified. Every tool page lists its sources and dates.
A to Z

Every tool, A to Z.