> For the complete documentation index, see [llms.txt](https://docs.meganova.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.meganova.ai/compute/serverless-models.md).

# Serverless Models

For text models, prices are calculated per 1 million tokens, including both input and output tokens. For image models, prices are based on image size and the default number of steps and will scale linearly with additional steps.

## Text Generation Models

### Pricing Structure

* **Per 1 million tokens** (input + output)
* **Tier-based pricing** with volume discounts
* **Real-time usage** - pay only for what you use

### Model Categories

#### Free Access Models (<100B parameters)

* **Manta Mini** - Available to all tiers
* **Manta Flash** - Available to Creator+ tiers
* **GLM-4.7-Flash** - Available to Creator+ tiers

#### Premium Models (>100B parameters)

* **Manta Pro** - Available to Legend tier
* **Nevoria** - Available to Legend tier
* **Mistral** - Available to Legend tier

## Image Generation Models

### Pricing Structure

* **Based on image size** and number of steps
* **Linear scaling** with additional steps
* **Tier-based access** to different model capabilities

### Image Sizes

* **Standard sizes**: 512x512, 1024x1024, 2048x2048
* **Custom sizes**: Available with higher tiers
* **Batch generation**: Discounts for multiple images

## Embedding Models

### Pricing Structure

* **Per 1 million tokens** for input text
* **Vector dimensions**: 384, 512, 1024, 1536
* **Batch processing**: Available for higher tiers

### Use Cases

* **Semantic search**
* **Content similarity**
* **Recommendation systems**
* **RAG (Retrieval Augmented Generation)**

## Tier-Specific Access

### Tier 1

* ✅ Free models only (<100B)
* ✅ Basic image generation
* ✅ Standard embedding models

### Tier 2

* ✅ Manta Flash, Manta Pro and GLM-4.7-Flash
* ✅ Enhanced image generation
* ✅ Advanced embedding capabilities

### Tier 3

* ✅ All Tier 2 features
* ✅ Priority model access
* ✅ Batch processing discounts

### Tier 4

* ✅ Access to >100B models
* ✅ Maximum image generation capabilities
* ✅ Enterprise-grade embedding models