> ## Documentation Index
> Fetch the complete documentation index at: https://docs.meigen.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# AI Model Comparison
> Compare AI image & video models on MeiGen — GPT Image 2.0, Seedream 5.0 Pro, Midjourney V8.1, Nano Banana 2, Seedream 5.0 Lite, Flux 2 Klein, Agnes Image 2.1 Flash, Seedance 2.0, Happyhorse 1.1, Veo 3.1, Agnes Video 2.0. Features, pricing, aspect ratios side by side.
MeiGen supports multiple AI models for image and video generation. Use the comparison table below to pick one, then scroll to its section for detailed specs.
## Quick Comparison
| Model | Credits | Speed | Ref Images | Best For |
| --------------------------- | ---------------------------------------------------------- | ---------------------- | ---------------------------------- | -------------------------------------------------------------------------- |
| **GPT Image 2.0** (default) | 2–80 (by resolution × quality) | Fast (\~35s) | 5 | Best accuracy & aesthetics, text rendering |
| **Nanobanana Pro** | 10 | Fast (\~30s) | 5 | High quality, complex scenes |
| **Nanobanana 2** | 5 | Fast (\~20s) | 5 | General use, wide ratio support |
| **Seedream 5.0 Pro** | 10 (1K) / 20 (2K) | Slow (\~2 min) | 10 | Multi-reference composition (up to 10 refs), character consistency |
| **Seedream 5.0 Lite** | 5 | Medium (\~42s) | 5 | Photorealism, 3K output |
| **Seedream 4.5** | 5 | Fast (\~32s) | 5 | Product photos, marketing |
| **Grok Imagine Quality** | 10 (1K) / 14 (2K) | Fastest (\~9s) | 3 | High-quality generation, text-to-image + image-to-image |
| **Midjourney V8.1** | 15 (1K) / 20 (2K) | Medium (\~50s) | 1 | Photorealistic, universal aesthetics |
| **Flux 2 Klein** | 2 | Fast (\~21s) | None | General-purpose base model, text rendering |
| **Z Image Turbo** | 2 | Fastest (\~11s) | None | Quick drafts, prototyping |
| **Agnes Image 2.1 Flash** | 1 | Fast (\~30s) | 1 | Cheapest model, high-volume drafts, covered by daily free credits |
| **Seedance 2.0** | Per-second (Mini 9–16/sec, Fast 13–20/sec, Pro 14–100/sec) | Video, \~2-6min | First/last frame + reference video | Three tiers (Mini / Fast / Pro), up to 4K on Pro, 4–15s, audio |
| **Grok Video 1.5** | Per-second (480p 16/sec, 720p 28/sec) | Video, \~35s | First frame (required) | First-frame image-to-video only, 4–15s, ratio follows input, audio |
| **Happyhorse 1.1** | Per-second (720p 18/sec, 1080p 28/sec) | Video, \~2-4min (720p) | First frame (single image) | 3–15s adjustable duration, text-to-video + first-frame i2v, audio included |
| **Veo 3.1** | Per generation (Fast 80–165, Pro 220–435) | Video, \~1-8min | Up to 2 (first/last frame) | Two tiers (Fast / Pro), 4/6/8s, 720p/1080p/4k same price, audio |
| **Agnes Video 2.0** | 10 per video (480p, 2/sec × 5s) | Video, \~2min | First frame (single image) | Fixed 5s / 480p, the only video model covered by daily free credits |
> GPT image 1.5 was retired on 2026-04-22 (replaced by GPT Image 2.0). Midjourney V7 was replaced by V8.1 on 2026-05-09.
## Image Models
### GPT Image 2.0
| | |
| -------------------- | ------------------------------------------------------------------- |
| **Model ID** | `gpt-image-2` |
| **Provider** | OpenAI |
| **Credits** | By resolution × quality — see table below |
| **Reference images** | Up to 5 |
| **Resolutions** | 1K, 2K, 4K |
| **Quality** | Standard / Medium / High |
| **Aspect ratios** | 1:1, 3:2, 2:3, 16:9, 9:16, 4:3, 3:4, 5:4, 4:5, 21:9, 9:21, 3:1, 1:3 |
**Credit matrix:**
| | Standard | Medium | High |
| ------ | -------- | ------ | ---- |
| **1K** | 2 | 10 | 18 |
| **2K** | 5 | 20 | 40 |
| **4K** | 10 | 32 | 80 |
All resolutions consume welcome or purchased credits (premium model — daily free credits don't apply). 4K is paid-only. Aspect ratio does not affect pricing — credits depend only on resolution × quality.
The current default model and MeiGen's **best overall image model**. Excels at instruction following, in-image text rendering, and complex multi-element scenes.
**Stricter moderation**: prompts involving real people, celebrities, brand logos, or violent/adult content may be rejected. Rephrase to drop specific names or try a different model.
***
### Nanobanana 2
| | |
| -------------------- | ----------------------------------------------------------------------- |
| **Model ID** | `nanobanana-2` |
| **Provider** | Google (Gemini) |
| **Credits** | 5 per image |
| **Reference images** | Up to 5 |
| **Resolutions** | 2K, 4K |
| **Aspect ratios** | 1:1, 3:4, 4:5, 2:3, 9:16, 1:4, 1:8, 4:3, 5:4, 3:2, 16:9, 21:9, 4:1, 8:1 |
The most affordable Gemini model with the widest range of aspect ratios, including ultra-wide (8:1) and ultra-tall (1:8) options. Good for quick iterations and exploration. 4K is paid-only.
**Content moderation**: Nanobanana 2 has stricter content filtering than other models. Prompts that mention **celebrities, brand names, real people, or face-swap scenarios** are more likely to be rejected. If your prompt is flagged, try rephrasing to avoid specific names or use a different model.
***
### Nanobanana Pro
| | |
| -------------------- | --------------------------------------------------- |
| **Model ID** | `gemini-3-pro-image-preview` |
| **Provider** | Google (Gemini) |
| **Credits** | 10 per image |
| **Reference images** | Up to 5 |
| **Resolutions** | 2K, 4K |
| **Aspect ratios** | 1:1, 3:4, 4:5, 2:3, 9:16, 4:3, 5:4, 3:2, 16:9, 21:9 |
The premium Gemini model with higher image quality. Better at complex scenes, fine details, and text rendering. Choose this when quality matters more than cost. 4K is paid-only.
***
### Seedream 5.0 Pro
| | |
| -------------------- | ----------------------------------------- |
| **Model ID** | `seedream-5.0-pro` |
| **Provider** | ByteDance |
| **Credits** | 10 (1K) / 20 (2K) per image |
| **Reference images** | Up to 10 |
| **Resolutions** | 1K, 2K — default 1K |
| **Aspect ratios** | 1:1, 3:4, 4:3, 16:9, 9:16, 3:2, 2:3, 21:9 |
ByteDance's flagship Seedream tier. Accepts **up to 10 reference images** — the highest of any model on MeiGen — making it the best pick for multi-subject composition and for keeping a character consistent across images. 2K is paid-only.
***
### Seedream 5.0 Lite
| | |
| -------------------- | ----------------------------------------- |
| **Model ID** | `seedream-5.0-lite` |
| **Provider** | ByteDance |
| **Credits** | 5 per image |
| **Reference images** | Up to 5 |
| **Resolutions** | 2K, 3K |
| **Aspect ratios** | 1:1, 3:4, 4:3, 16:9, 9:16, 3:2, 2:3, 21:9 |
ByteDance's latest lightweight model. Offers good quality at low cost with optional 3K resolution output. Strong at photorealistic content and Chinese-style aesthetics. 3K is paid-only.
***
### Seedream 4.5
| | |
| -------------------- | ----------------------------------------- |
| **Model ID** | `seedream-4.5` |
| **Provider** | ByteDance |
| **Credits** | 5 per image |
| **Reference images** | Up to 5 |
| **Resolutions** | 2K, 4K |
| **Aspect ratios** | 1:1, 4:3, 3:4, 16:9, 9:16, 3:2, 2:3, 21:9 |
The previous generation Seedream model. Still produces solid results, particularly good for product photography and marketing visuals. 4K is paid-only.
***
### Grok Imagine Quality
| | |
| -------------------- | ----------------------------------- |
| **Model ID** | `grok-image` |
| **Provider** | xAI |
| **Credits** | 10 (1K) / 14 (2K) per image |
| **Reference images** | Up to 3 |
| **Resolutions** | 1K, 2K — default 1K |
| **Aspect ratios** | 1:1, 3:2, 2:3, 4:3, 3:4, 16:9, 9:16 |
xAI's high-quality image generation model. Supports both text-to-image and image-to-image (up to 3 reference images for editing or guided generation). 2K is paid-only.
***
### Midjourney V8.1
| | |
| -------------------- | -------------------------------------------- |
| **Model ID** | `midjourney-v8.1` |
| **Provider** | Midjourney |
| **Credits** | 15 (1K) / 20 (2K) per generation |
| **Reference images** | 1 (content or style reference) |
| **Resolutions** | 1K (1024×1024) / 2K (2048×2048) — default 1K |
| **Aspect ratios** | 1:1, 4:3, 3:2, 16:9, 3:4, 2:3, 9:16 |
| **Quality** | Standard / High — default High (same price) |
Midjourney's photorealistic flagship. Returns 4 candidate images per generation (charged once). Choose **2K** for native high-resolution output, **High quality** for finer detail at no extra cost. 2K is paid-only.
**Features:**
* Content and style reference modes
* Advanced parameters (stylize, chaos, raw, etc.) for fine-tuning the output
* Auto-translates non-English prompts to English
* Native 2K output toggle
* Quality (Standard / High) toggle
***
### Flux 2 Klein
| | |
| ---------------------- | ----------------------------------------- |
| **Model ID** | `flux2-klein` |
| **Provider** | Black Forest Labs |
| **Credits** | 2 per image |
| **Daily free credits** | Eligible |
| **Reference images** | Not supported |
| **Aspect ratios** | Auto, 1:1, 4:3, 3:4, 16:9, 9:16, 3:2, 2:3 |
A base model from Black Forest Labs, suited to everyday general-purpose creation.
***
### Z Image Turbo
| | |
| ---------------------- | --------------------------------------------------------- |
| **Model ID** | `z-image-turbo` |
| **Provider** | Alibaba (Qwen) |
| **Credits** | 2 per image |
| **Daily free credits** | Eligible |
| **Reference images** | Not supported |
| **Resolutions** | Fixed |
| **Aspect ratios** | 1:1, 2:3, 3:2, 3:4, 4:3, 7:9, 9:7, 9:16, 9:21, 16:9, 21:9 |
The fastest and cheapest model — ideal for quick drafts when you need results in seconds.
**No reference image support.** Z Image Turbo generates images from text only — you cannot upload reference images with this model. If you need to use reference images, choose a different model.
***
### Agnes Image 2.1 Flash
| | |
| ---------------------- | ----------------------------------- |
| **Model ID** | `agnes-image-2.1-flash` |
| **Provider** | Agnes |
| **Credits** | 1 per image |
| **Daily free credits** | Eligible |
| **Reference images** | 1 |
| **Aspect ratios** | 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3 |
The cheapest model on MeiGen at 1 credit per image, and the only basic-tier **image** model that accepts a reference image. Good for high-volume drafting and low-cost image-to-image iteration.
***
## Video Models
### Seedance 2.0
| | |
| -------------------- | --------------------------------------------------------- |
| **Model ID** | `seedance-2-0` |
| **Provider** | ByteDance |
| **Credits** | Per-second pricing (see below) |
| **Duration** | 4-15 seconds (default 4) |
| **Reference images** | First/last frame (image-to-video) |
| **Reference video** | All tiers (Pro 480p has no With-reference-video rate) |
| **Resolutions** | Mini and Fast: 480p / 720p; Pro: 480p / 720p / 1080p / 4K |
| **Aspect ratios** | adaptive, 16:9, 4:3, 1:1, 3:4, 9:16, 21:9 |
| **Audio** | Auto-generated |
ByteDance's latest video generation model. Supports text-to-video, image-to-video (first/last frame), and video continuation. All outputs include AI-generated audio. The `adaptive` aspect ratio automatically matches the reference image/video dimensions. Seedance 2.0 is also the only video model with **character assets**: save a character once, then reference it with `@` in your prompt to keep the same subject across shots.
**Three tiers**, picked from the tier switch (Mini / Fast / Pro) in the video sidebar. Same Seedance 2.0 model, same params, just different rendering quality and price:
* **Mini** (default): cheapest and most lightweight; 480p / 720p, supports reference-video continuation
* **Fast**: quick turnaround at 480p / 720p
* **Pro**: higher fidelity, with **native 1080p and 4K** rendering
#### Pricing
Per-second. Rate depends on **tier** (Mini / Fast / Pro) × **whether a reference video is uploaded**:
| Resolution | Mini Direct | Mini With reference video | Fast Direct | Fast With reference video | Pro Direct | Pro With reference video |
| ---------- | ----------- | ------------------------- | ----------- | ------------------------- | ---------- | ------------------------ |
| 480p | 9/sec | 7/sec | 13/sec | 8/sec | 14/sec | — |
| 720p | 16/sec | 13/sec | 20/sec | 14/sec | 22/sec | 16/sec |
| 1080p | — | — | — | — | 42/sec | 28/sec |
| 4K | — | — | — | — | 100/sec | — |
Mini and Fast only support 480p / 720p. Pro adds native 1080p and 4K (4K is direct-generation only). All three tiers support reference-video continuation; Pro has no 480p With-reference-video rate.
**With-reference-video billing rule (all tiers):**
`billable seconds = max(reference video duration + your selected duration, min billable)`
Output length is always your selected duration (4–15), independent of the reference video's length.
| Your selected duration (sec) | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 |
| ---------------------------- | - | - | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- |
| Min billable (sec) | 7 | 9 | 10 | 12 | 14 | 15 | 17 | 19 | 20 | 22 | 24 | 25 |
#### Examples
| Scenario | Calculation | Total |
| ----------------------------------------------- | ---------------------------- | ------- |
| Mini 480p × 4s direct | 9 × 4 | **36** |
| Mini 720p × 5s direct | 16 × 5 | **80** |
| Fast 480p × 5s direct | 13 × 5 | **65** |
| Fast 720p × 5s direct | 20 × 5 | **100** |
| Fast 720p, 5s reference + 10s selected duration | 14 × max(5+10, 17) = 14 × 17 | **238** |
| Pro 720p × 5s direct | 22 × 5 | **110** |
| Pro 1080p × 5s direct | 42 × 5 | **210** |
| Pro 4K × 5s direct | 100 × 5 | **500** |
| Pro 1080p, 5s reference + 10s selected duration | 28 × max(5+10, 17) = 28 × 17 | **476** |
If generation fails, the **full** credit amount is automatically refunded — including all per-second charges.
***
### Grok Video 1.5
| | |
| -------------------- | ----------------------------------------------------------------------- |
| **Model ID** | `grok-video` |
| **Provider** | xAI |
| **Credits** | Per-second (see below) |
| **Duration** | 4-15 seconds (default 4) |
| **Reference images** | 1 — first frame **required** (image-to-video only) |
| **Reference video** | Not supported |
| **Resolutions** | 480p, 720p (default 480p) |
| **Aspect ratios** | Follows the first-frame image automatically (no manual ratio selection) |
| **Audio** | Auto-generated |
xAI's image-to-video model. A starting (first-frame) image is **required** — pure text-to-video is not supported. The output aspect ratio always follows your input image, and AI-generated audio is included.
#### Pricing
| Resolution | Rate |
| ---------- | -------------- |
| 480p | 16 credits/sec |
| 720p | 28 credits/sec |
**Examples:**
| Scenario | Calculation | Total |
| ---------- | ----------- | ------- |
| 480p × 4s | 16 × 4 | **64** |
| 480p × 10s | 16 × 10 | **160** |
| 720p × 4s | 28 × 4 | **112** |
| 720p × 10s | 28 × 10 | **280** |
If generation fails, the **full** credit amount is automatically refunded.
***
### Happyhorse 1.1
| | |
| -------------------- | ------------------------------- |
| **Model ID** | `happyhorse-1.1` |
| **Provider** | Alibaba |
| **Credits** | Per-second (see below) |
| **Duration** | 3–15 seconds (default 4) |
| **Reference images** | 1 (first frame, image-to-video) |
| **Reference video** | Not supported |
| **Resolutions** | 720p, 1080p (default 720p) |
| **Aspect ratios** | Auto, 16:9, 9:16, 1:1, 4:3, 3:4 |
| **Audio** | Auto-generated |
Alibaba's video model — text-to-video and single-image first-frame image-to-video. AI-generated audio included. No reference-video continuation. **Auto** aspect ratio infers the output ratio from your prompt or reference image.
Typical 720p runs take about 2–4 minutes; the time scales with the clip length you choose, and first-frame image-to-video generally finishes faster than pure text-to-video. 1080p takes noticeably longer.
#### Pricing
| Resolution | Rate |
| ---------- | -------------- |
| 720p | 18 credits/sec |
| 1080p | 28 credits/sec |
**Examples:**
| Scenario | Calculation | Total |
| ----------- | ----------- | ------- |
| 720p × 5s | 18 × 5 | **90** |
| 720p × 10s | 18 × 10 | **180** |
| 1080p × 5s | 28 × 5 | **140** |
| 1080p × 10s | 28 × 10 | **280** |
If generation fails, the **full** credit amount is automatically refunded.
***
### Veo 3.1
| | |
| -------------------- | ------------------------------- |
| **Model ID** | `veo-3.1` |
| **Provider** | Google |
| **Credits** | Per generation (see below) |
| **Duration** | 4, 6, or 8 seconds (default 4) |
| **Reference images** | Up to 2 (as first/last frame) |
| **Reference video** | Not supported |
| **Resolutions** | 720p, 1080p, 4k (default 1080p) |
| **Aspect ratios** | Auto, 16:9, 9:16 |
| **Audio** | Auto-generated (always on) |
Google's video generation model. Creates short clips with AI-generated audio synced to the prompt. **Auto** aspect ratio infers the output ratio from your prompt or reference image. Pick this for premium visual quality with native audio; for longer durations (>8s) or wider ratio support, use Seedance 2.0.
**Two tiers** — pick one from the tier switch (Fast / Pro) in the video sidebar. Same Veo 3.1 model, same params, just different rendering quality and price:
* **Fast** (default) — quick turnaround, balanced quality
* **Pro** — higher fidelity, longer rendering time
#### Pricing
Per generation. Price depends on **tier** (Fast / Pro) × **duration** (4 / 6 / 8s). **All resolutions (720p / 1080p / 4k) share the same price** — picking a higher resolution costs no extra credits. Two caveats: 4k requires a paid account, and it takes noticeably longer to render.
| Duration | Fast | Pro |
| -------- | ---- | --- |
| 4s | 80 | 220 |
| 6s | 125 | 330 |
| 8s | 165 | 435 |
#### Examples
| Scenario | Total |
| ------------------------ | ------- |
| Fast 4s (any resolution) | **80** |
| Fast 8s (any resolution) | **165** |
| Pro 4s (any resolution) | **220** |
| Pro 8s 4k | **435** |
If generation fails, the **full** credit amount is automatically refunded.
***
### Agnes Video 2.0
| | |
| ---------------------- | ------------------------------- |
| **Model ID** | `agnes-video-2.0` |
| **Provider** | Agnes |
| **Credits** | 10 per video |
| **Daily free credits** | Eligible |
| **Duration** | 5 seconds (fixed) |
| **Reference images** | 1 (first frame, image-to-video) |
| **Reference video** | Not supported |
| **Resolutions** | 480p |
| **Aspect ratios** | Auto, 16:9, 9:16, 1:1, 4:3, 3:4 |
The only basic-tier video model — and the only one your daily free credits cover. Supports text-to-video and single-image first-frame image-to-video at a fixed 5 seconds / 480p. For higher resolution or longer clips, use Seedance 2.0, Happyhorse 1.1, or Veo 3.1.
#### Pricing
| Resolution | Rate |
| ---------- | ------------- |
| 480p | 2 credits/sec |
Duration is fixed at 5 seconds, so every generation costs **10 credits**.
If generation fails, the **full** credit amount is automatically refunded.
***
## Known Limitations
This model is more sensitive to content involving real people, celebrities, brand logos, and face-swap requests. Prompts that work on other models may be rejected here. If you get a content violation error, rephrase your prompt to remove specific names or try a different model.
Both base models generate images from text only. The reference image upload section will be hidden when either is selected. If you need to use reference images, switch to any other model.
Midjourney V8.1 accepts only 1 reference image, while most other models accept up to 5 (Seedream 5.0 Pro accepts up to 10). In exchange it offers more control over how the reference is used (content vs. style reference).
Unlike other models that return a single image, Midjourney V8.1 returns 4 candidate images per generation. All candidates are saved and can be browsed in the image detail dialog. You are only charged once (15 credits at 1K / 20 credits at 2K) for all 4 images.
Unlike Seedance 2.0, Happyhorse does not support video continuation. Reference image input is single first-frame only. For continuation, use Seedance 2.0.
GPT Image 2.0 applies stricter moderation than other models to real people, celebrities, brand logos, and adult/violent content. Rephrase your prompt to drop specific names, or try a different model.
Seedance bills **per second**, not per image — and uploading a reference video uses a separate rate table with a minimum-billable floor. See the **Seedance 2.0** section above for the full rates and formula.
The Mini tier (the default) is the lightweight option, capped at 720p (it does support reference-video continuation). For 1080p or 4K output, switch to **Pro**.
Grok Video 1.5 is image-to-video only — you **must** upload a starting (first-frame) image. Pure text-to-video is not supported. The output aspect ratio follows your input image. If you want text-to-video, use Seedance 2.0, Happyhorse 1.1, or Veo 3.1.
Grok Video 1.5 outputs at 480p or 720p only. For higher resolutions (1080p / 4k), use Seedance 2.0 or Veo 3.1.
Veo 3.1 only supports 16:9 / 9:16 (plus Auto). Square (1:1), 4:3 / 3:4 portrait, and other ratios are not available. For wider ratio support, use Seedance 2.0 or Happyhorse 1.1.
720p / 1080p / 4k all cost the same on Veo 3.1, so 4k costs no extra credits. Two things to know: 4k is **locked for free accounts** (selecting it opens the upgrade dialog), and rendering takes longer — 4k typically takes 6–8 minutes versus 1–4 minutes for 720p / 1080p. The same 4k lock applies to Seedance 2.0 Pro.