AI Models

Gab AI gives you access to the world's leading AI models through a single interface. Learn about each model's strengths and when to use them.

How Model Selection Works

By default, Gab AI uses intelligent routing to automatically select the best model for your request. You can also manually select a model using the model selector in the chat interface.

Default AI is Free in the App

Our default AI model is free for Plus subscribers when chatting in Gab AI. Not included in API — API requests always use credits. Other third-party models use credits based on usage.

Text & Chat Models

These models power conversations, writing, coding, and analysis:

Arya 3

Gab AI

Gab AI's core model. Free for Plus subscribers in the app (not included in api). Strengths: General conversation, Writing assistance, Value-aligned responses

Claude Opus 4.8

Anthropic

Anthropic's most capable model with frontier reasoning and a step-change improvement in agentic coding. Strengths: Agentic coding, Complex reasoning, Analysis

Claude Opus 4.7

Anthropic

Premium model combining maximum intelligence with practical performance. Strengths: Complex reasoning, Analysis, Creative writing

Claude Sonnet 4.6

Anthropic

Hybrid Sonnet excels at coding, data analysis, and planning with fast, efficient responses. Strengths: Coding, Data analysis, Planning

Claude Sonnet 4.5

Anthropic

Balanced Sonnet model strong at coding, data analysis, and everyday reasoning. Strengths: Coding, Data analysis, Reasoning

Claude Haiku 4.5

Anthropic

Fastest, most cost-efficient Anthropic model. Strengths: Speed, Efficiency, Quick tasks

GPT-5.5

OpenAI

OpenAI's flagship general-purpose model with strong reasoning and tool use. Strengths: General reasoning, Tool use, Writing

GPT-5.3 Codex

OpenAI

OpenAI's most advanced agentic coding model, combining frontier software engineering performance with broader reasoning. Strengths: Code generation, Software engineering, Reasoning

o3

OpenAI

OpenAI's deep reasoning model for math, science, and complex multi-step problems. Strengths: Deep reasoning, Math, Science

Gemini 3.1 Pro

Google

One of the best models in the world for multimodal understanding, and Google's most powerful agentic and vibe-coding model. Strengths: Multi-modal, Agentic, Coding

Gemini 3.5 Flash

Google

Fast, balanced Gemini model strong at reasoning, coding, and everyday multimodal tasks. Strengths: Reasoning, Coding, Speed

Gemini 3.1 Flash Lite

Google

Lightweight Gemini model tuned for low-latency responses and real-time web access. Strengths: Speed, Web search, Cost-efficient

DeepSeek V4 Pro

DeepSeek

DeepSeek's top-tier model for advanced reasoning, math, and coding tasks. Strengths: Reasoning, Coding, Math

DeepSeek V4 Flash

DeepSeek

Faster, cost-efficient DeepSeek variant for everyday text and code generation. Strengths: Speed, Cost-efficient, Coding

GLM 5.1

Z.ai

Z.ai's flagship model — strong text generation and coding with multilingual support. Strengths: Text generation, Coding, Multilingual

Qwen 3.7 Max

Alibaba

Alibaba's top-tier Qwen model for advanced reasoning, coding, and agentic tasks. Strengths: Reasoning, Coding, Agentic tasks

Qwen 3.7 Plus

Alibaba

Cost-effective Qwen3.7 model with vision-language understanding, coding, and agentic tool use. Strengths: Vision-language, Coding, Cost-efficient

Qwen 3.5 397B Thinking

Alibaba

Thinking variant of Qwen 3.5 397B — strengthens chain-of-thought for harder analytical tasks. Strengths: Advanced reasoning, Math, Problem-solving

Kimi K2.6

Moonshot

Moonshot AI's next-generation multimodal model for long-horizon coding, UI/UX generation, and multi-agent orchestration. Strengths: Long-horizon coding, Multimodal, Agent orchestration

Kimi K2 Thinking

Moonshot

Moonshot AI's most advanced open reasoning model, extending the K2 series into agentic, long-horizon reasoning. Strengths: Agentic reasoning, Long-horizon tasks, Open source

MiniMax M3

MiniMax

Latest MiniMax model — strong coding and reasoning with efficient inference cost. Strengths: Coding, Reasoning, Cost-efficient

Hermes 4 405B

Nous Research

The largest model in the Hermes 4 family — the fullest expression of Nous Research design, focused on advanced reasoning. Strengths: Advanced reasoning, Instruction following, Open source

Sonar Reasoning

Perplexity

Chain-of-thought model built for rapid problem solving with real-time web access. Strengths: Web search, Current events, Research

Sonar

Perplexity

Fast Perplexity model with grounded answers backed by real-time web search. Strengths: Web search, Current events, Speed

Gemma 4 26B

Google

Open-weights Gemma model — capable general-purpose chat with strong instruction following. Strengths: Open-weights, Instruction following, Cost-efficient

Image Generation Models

Create stunning AI-generated images with these models:

GPT Image 2

OpenAI

OpenAI's state-of-the-art image generation and editing model with high-fidelity image inputs and flexible sizes.

GPT Image 1 Mini

OpenAI

Faster, cost-efficient OpenAI image model that accepts both text and image inputs.

Nano Banana 2

Google

Google's fast, high-quality image generation and editing model.

Seedream 4.5

ByteDance

Latest Seedream — unified image generation and editing with sharper text rendering.

Seedream 4.0

ByteDance

Integrates image generation and image editing capabilities in a single, unified architecture.

Qwen Image 2

Alibaba

Latest Qwen image model — strong text rendering and precise instruction following.

Qwen Image

Alibaba

A powerful image generation model capable of complex text rendering and precise image editing.

Wan 2.7 Image

Alibaba

Alibaba WAN image model for high-quality generation and editing from text or reference images.

Image Generator

Gab AI

Generate vivid, realistic images based on a text prompt.

Video Generation

Generate AI videos from text descriptions:

Veo 3.1

Google

Google's state-of-the-art model for generating high-fidelity videos.

Veo 3.1 Fast

Google

Faster and more cost effective version of Google's Veo 3.1.

Veo 3.1 Img2Vid

Google

Google's state-of-the-art model for generating high-fidelity videos from images.

Kling 3 Pro

Kling

Top-tier motion video generation with exceptional prompt precision.

Kling 3.0 Img2Vid

Kling

Top-tier motion video generation from images with exceptional prompt precision.

Kling 2.5 Turbo Pro

Kling

Top-tier video generation with exceptional prompt precision.

Kling 2.5 Turbo Pro Img2Vid

Kling

Top-tier video generation from images with exceptional prompt precision.

Hailuo 2.3 Pro

MiniMax

Advanced text-to-video generation model with 1080p resolution.

Hailuo 2.3

MiniMax

Advanced text-to-video generation model with 1080p resolution.

Wan 2.7

Alibaba

Latest Alibaba WAN — generates high-quality AI videos with audio from text or images.

Wan 2.5

Alibaba

Generate high-quality AI videos with audio using Alibaba WAN 2.5.

Seedance 2.0

ByteDance

ByteDance video model with strong motion coherence for both text-to-video and image-to-video.

Music Generation

Create original music and songs with AI:

MiniMax Music 2.6

MiniMax

Latest MiniMax music model — generates high-quality, diverse musical compositions from text prompts.

MiniMax Music 2.0

MiniMax

Generate high-quality, diverse musical compositions from text prompts.

3D Generation

Turn text prompts or images into downloadable 3D models:

Hunyuan 3D v3.1 Pro

Tencent

High-fidelity 3D model generation from a text prompt or an image.

Hunyuan 3D v3.1 Rapid

Tencent

Faster Hunyuan 3D generation for quick iterations from text or an image.

Tripo H3.1

Tripo3D

Generate detailed, production-ready 3D meshes from a text prompt.

Meshy v6

Meshy

Create textured 3D models from text or an image.

Talking Avatars

Generate lifelike talking-avatar videos from a portrait, script, or audio:

HeyGen Avatar 4

HeyGen

Generate lifelike talking-avatar videos from a portrait and a script.

HeyGen Video Agent

HeyGen

Create talking-avatar videos driven by your text or audio.

Kling Avatar v2 Pro

Kling

High-quality talking and singing avatar videos from an image and audio.

Hunyuan Avatar

Tencent

Animate a portrait into a talking avatar driven by audio.

MultiTalk Avatar

Fal AI

Lip-synced talking-avatar video generation from audio or text-to-speech.

Speech & Sound Effects

Convert text to natural speech and generate sound effects:

GPT-4o mini TTS

OpenAI

Natural text-to-speech with a range of expressive voices.

ElevenLabs Sound Effects V2

ElevenLabs

Generate sound effects and short audio clips from a text description.

CassetteAI Sound Effects

CassetteAI

Fast text-to-sound-effect generation.

Model Availability

We're constantly adding new models as they become available. Model availability may vary based on your subscription tier:

Free Tier

Limited access to basic models

Plus

All models + free default model messages + 2,000 credits/mo

Enterprise

Custom model access + volume pricing

Default AI is Free in the App

Our default AI model is free for Plus subscribers when chatting in Gab AI. Not included in API — API requests always use credits. Other third-party models use credits based on usage.