ZMIME
Prompt library · Updated 2026

Prompts for people who ship the work.

A working library of prompt templates, organised by what you are making — text, images, video, audio, avatars, software, marketing — with separate packs for the tools you actually open.

  • 07Categories
  • 08Tool packs
  • 73Templates
  • 00Signup walls
01 — AI models

Top 24 AI models right now

Current language models released in the last 240 days, ranked by published Epoch Capabilities Index (ECI) scores. Equal scores share a rank; there are no estimated scores or lab quotas. Higher ECI means stronger measured capability, not necessarily better value for your task. Image and video rankings use blind votes from Arena’s text-to-image and text-to-video leaderboards. Models released in the last 60 days are tagged New. Open any model for prices, specs and head-to-head comparisons.

Recently released AI models

Released in the last 30 days, newest first. Benchmark scores may not yet be available.

  1. NewGPT-6.2 AstraOpenAIReleased
  2. NewGPT-6.2 SolOpenAIReleased
  3. NewMAI-Image-2.7MicrosoftReleased
  4. NewMAI-Image-2.7 FlashMicrosoftReleased
  5. NewHy4 flashTencentReleased
  6. NewQwen4 MaxAlibaba (Qwen)Released
  7. NewGemini 4 HydrogenGoogleReleased
  8. NewGemini 4 CarbonGoogleReleased
  9. NewClaude Fable 5.2AnthropicReleased
  10. NewMuse EmberMetaReleased
  11. NewGemini Omni 1.2 FlashGoogleReleased
  12. NewGPT-6 CyberOpenAIReleased
  13. NewInkling 2 SmallThinking MachinesReleased
  14. NewLing 3.2 FlashinclusionAIReleased
  15. NewGPT Image 2.6 FlareOpenAIReleased
  16. NewGPT Image 2.6 SunburstOpenAIReleased
  17. NewHy4TencentReleased
  18. NewLongCat-2.5Meituan (LongCat)Released
  19. NewParticle-1Particle AiReleased
  20. NewGPT-6.1 TerraOpenAIReleased
  21. NewKimi K3.1 ProMoonshot AIReleased
  22. NewFlintRecursiveReleased
  23. NewOrnith 1.5 122BDeepReinforceReleased
  24. NewMiniMax-M3.1-FlashMiniMaxReleased

Top language models

  1. Rank 01NewGPT-6.2 AstraOpenAIECI 174.2
  2. Rank 02NewClaude Fable 5.2AnthropicECI 173.5
  3. Rank 03NewGemini 4 CarbonGoogleECI 169.8
  4. Rank 04GPT-6 AstraOpenAIECI 169.2
  5. Rank 05NewClaude Opus 5.5AnthropicECI 168.7
  6. Rank 06NewGPT-6.1 SolOpenAIECI 168.6
  7. Rank 07NewGPT-6.1 TerraOpenAIECI 167.7
  8. Rank 08Claude Fable 5.1AnthropicECI 167.1
  9. Rank 09NewClaude Sonnet 5.5AnthropicECI 166.6
  10. Rank 10NewGemini 4 ArgonGoogleECI 166.6
  11. Rank 11Claude Opus 5AnthropicECI 164.7
  12. Rank 12Claude Fable 5AnthropicECI 164.2
  13. Rank 13NewGPT-6 SolOpenAIECI 163.8
  14. Rank 14GPT-5.6 SolOpenAIECI 163.1
  15. Rank 15GPT-5.5 ProOpenAIECI 162.8
  16. Rank 16NewMuse WonderMetaECI 161.7
  17. Rank 17NewKimi K3.1 ProMoonshot AIECI 161.4
  18. Rank 18GPT-5.6 TerraOpenAIECI 160.8
  19. Rank 19NewClaude Haiku 5.5AnthropicECI 160.3
  20. Rank 20GPT-5.5OpenAIECI 160.2
  21. Rank 21Claude Opus 4.8AnthropicECI 159.9
  22. Rank 22NewDeepSeek V4.1 Flash 1017DeepSeekECI 159.4
  23. Rank 23Gemini 3.8 FlashGoogleECI 158.8
  24. Rank 24Kimi K3Moonshot AIECI 158.6

View all 523 AI models

Top text-to-image models

  1. GPT Image 2.5 SunburstOpenAIArena 1,427
  2. GPT Image 2.5 FlareOpenAIArena 1,401
  3. GPT Image 2OpenAIArena 1,381
  4. NewMAI-Image-2.6MicrosoftArena 1,329
  5. NewNano Banana 2.1GoogleArena 1,325
  6. Grok Imagine Image 2.0xAIArena 1,317
  7. Reve 2.1reveArena 1,300
  8. Muse ImageMetaArena 1,269

View all 144 image AI models

Top text-to-video models

  1. NewGemini Omni 1.2 FlashGoogleArena 1,540
  2. Gemini Omni 1.1 FlashGoogleArena 1,513
  3. Gemini Omni Flash PreviewGoogleArena 1,509
  4. FLUX 3 VideoBlack Forest LabsArena 1,491
  5. Grok Imagine Video 1.5xAIArena 1,488
  6. Seedance 2.0ByteDance SeedArena 1,483
  7. Seedance 2.5ByteDance SeedArena 1,479
  8. Wan v3.0 VideoAlibaba (Qwen)Arena 1,476

View all 82 video AI models

04 — Method

Seven parts, every time

Medium and model change the details, not the skeleton. When output disappoints, one of these is missing.

  1. Role and audience

    Who produces this, and who reads it. Sets vocabulary and depth.

  2. One task

    A single deliverable. Anything with an "and then" belongs in a second prompt.

  3. Context you supply

    Source text, product facts, code, screenshots. Never rely on recall for something you can paste.

  4. Constraints

    Length, format, tone, banned words, technical limits. Most of the quality lives here.

  5. Output shape

    Headings, a table, JSON, a shot list. Name it rather than hoping for it.

  6. Uncertainty rule

    How to flag doubt, and what to do when the answer is not in the material.

  7. Scope fence

    What to leave alone. Essential once a model can edit files or run commands.

Fast diagnostic: read your prompt as a competent stranger would. Every question you would need to ask is a gap the model fills by guessing.

Pick a medium and get going

No account, no paywall, no newsletter gate. Every template is copy-and-go, and every page explains why it is shaped the way it is.