Every AI Model on Promptha: LLMs, Image, Video, Audio & More

Promptha gives you access to 80+ AI models from leading providers—all through a single interface. No separate accounts. No managing multiple API keys. Just pick the best model for your task and go.

This guide covers every model available, organized by category. Whether you need text generation, image creation, video production, or audio synthesis, you'll find the right model here.

Table of Contents

1. LLM Models (Text Generation)
2. Image Generation Models
3. Video Generation Models
4. Audio & Music Models
5. 3D Generation Models
6. Utility Models
7. How to Choose the Right Model
8. Model Tiers Explained

---

LLM Models (Text Generation)

Large Language Models handle text generation, analysis, coding, and reasoning. Promptha offers models from four major providers.

OpenAI Models

GPT-4o is OpenAI's flagship. It handles text and images, follows instructions precisely, and excels at coding. Use GPT-4o Mini for simpler tasks where speed and cost matter more than maximum capability.

Anthropic Models (Claude)

Claude Sonnet 4 is Anthropic's newest and most capable model. Claude models excel at nuanced writing, careful analysis, and following complex instructions. The 200K context window means they can process entire books or large codebases at once.

Google Models (Gemini)

Gemini 3 Flash is Google's latest. The standout feature is the 1 million token context window—you can analyze entire codebases, book series, or video transcripts in a single prompt. Gemini models also understand images, video, and audio natively.

DeepSeek Models

DeepSeek offers excellent value. These models perform well on reasoning and coding tasks at a fraction of the cost of premium models. Great for high-volume applications or when budgets are tight.

---

Image Generation Models

Promptha connects to the best image generation models through Fal.ai and Replicate.

Flux Family (Black Forest Labs)

Flux models are the current leaders in image quality. Flux 2 Pro produces the best results, while Flux Schnell is blazing fast for quick iterations. Use Flux Kontext when you need consistent characters across multiple generations.

Ideogram

Ideogram excels at rendering text in images. If you need logos, posters, or graphics with readable text, Ideogram is often the best choice.

Stable Diffusion

Stable Diffusion models are reliable workhorses. SDXL is open-source and versatile—great for experimentation and when you need fine-tuned control.

Google Imagen

Imagen 4 from Google offers exceptional prompt understanding. It follows complex instructions well and produces high-quality photorealistic images.

Other Image Models

---

Video Generation Models

Video AI has advanced rapidly. Promptha offers the latest models from multiple providers.

OpenAI Sora

Sora 2 is OpenAI's flagship video model. It creates high-quality, realistic videos with complex scenes and motion. Use Sora 2 I2V to animate still images.

Google Veo

Veo 3.1 from Google understands physics and motion exceptionally well. It produces high-fidelity videos with natural movement. Use the Fast variant for quick iterations.

Kling (Kuaishou)

Kling excels at maintaining character consistency across video frames. Great for character animations and story-based content.

LTX (Lightricks)

LTX-2 is the budget-friendly option. When you need video quickly without premium cost, LTX delivers good results.

Other Video Models

---

Audio & Music Models

Generate speech, music, and sound effects with specialized audio models.

Text-to-Speech

Maya TTS produces expressive, natural-sounding speech. Chatterbox TTS is great for entertainment—memes, games, AI agents. Speech 02 HD offers premium professional quality.

Music Generation

Lyria 2 from Google is the premium choice for AI-composed music. Beatoven Music generates royalty-free instrumentals for videos and podcasts.

Sound Effects

Beatoven SFX creates sound effects for games, videos, and multimedia projects.

---

3D Generation Models

Create 3D models from text or images.

Rodin generates 3D models from text descriptions or images—useful for game assets, product visualization, and 3D printing. SAM 3 variants reconstruct accurate 3D models from photographs.

---

Utility Models

Specialized models for specific tasks.

Upscaling

Crystal Upscaler increases image resolution while preserving detail and color fidelity.

Background Removal

Bria Background Remove extracts subjects from images with high accuracy—essential for e-commerce and product photography.

Avatar & Lipsync

Creatify Aurora generates talking avatar videos from text. Sync Lipsync V2 synchronizes lip movements to audio—useful for dubbing and localization.

---

How to Choose the Right Model

For Text Generation

1. General tasks with images: GPT-4o or Claude Sonnet 4
2. Long documents (100K+ tokens): Gemini models (1M context)
3. Budget-conscious: GPT-4o Mini, Claude 3.5 Haiku, DeepSeek Chat
4. Coding focus: Claude Sonnet 4, DeepSeek Coder
5. Deep analysis: Claude 3 Opus, Gemini 2.5 Pro

For Image Generation

1. Maximum quality: Flux 2 Pro, Flux Pro 1.1 Ultra
2. Text in images: Ideogram V3
3. Fast iterations: Flux Schnell, Ideogram V3 Turbo
4. Photorealistic: Flux Realism, Imagen 4
5. Vector/design: Recraft V3, Recraft V3 SVG
6. Consistent characters: Flux Kontext

For Video Generation

1. Premium quality: Sora 2, Veo 3.1
2. Character consistency: Kling Video 2.6
3. Budget-friendly: LTX-2
4. Image animation: Sora 2 I2V, Veo 3.1 I2V, Pika 2.2
5. Quick iterations: Veo 3.1 Fast, LTX-2

For Audio

1. Professional voiceover: Maya TTS, Speech 02 HD
2. Fun/games: Chatterbox TTS
3. Background music: Beatoven Music, MiniMax Music V2
4. Premium compositions: Lyria 2

---

Model Tiers Explained

Promptha organizes models into three tiers:

Premium Tier

• Highest quality output
• Best for production work
• Higher cost per generation
• Examples: Flux 2 Pro, Sora 2, Claude Sonnet 4

Standard Tier

• Good quality-to-cost ratio
• Suitable for most tasks
• Balanced performance
• Examples: Ideogram V3, Kling Video, Gemini Flash

Budget Tier

• Fastest generation
• Lowest cost
• Great for iterations and testing
• Examples: Flux Schnell, LTX-2, GPT-4o Mini

Use premium models for final output. Use budget models for exploration and iteration. Standard models work well when you need both quality and volume.

---

Using Models in Promptha

In Fabrics

When you use a Fabric, the creator has already selected the optimal model for that task. You can often switch models in the settings if you prefer a different option.

In Ask

Ask assistants support dynamic model switching. Start with one model, switch to another mid-conversation based on what you need.

In AskGL

AskGL gives you direct model control with the @provider.model syntax:

/image @fal.flux-schnell sunset over mountains
/write @anthropic.claude-sonnet-4 blog post about AI
/video @fal.veo-3.1 cinematic drone shot

---

What's Next?

Now that you know what models are available:

• What is AskGL? - Control models with command syntax
• What is a Fabric? - Use pre-configured AI tools
• What is Ask? - Conversational AI assistants
• Claude vs GPT-4 vs Gemini - LLM comparison
• Image Generation Models Compared - Deep dive on image AI

Promptha brings together the best AI models in one platform. Instead of managing multiple accounts and APIs, you access everything through a unified interface. Pick the right model for each task, and let the platform handle the rest.