AI FeaturesGetting Started

Choosing an AI Model

Compare AI models available in Firebuzz and learn which one to use for your tasks.

Firebuzz gives you access to 12 AI models from leading providers. Each model has different strengths in terms of cost, speed, and intelligence. Choosing the right model helps you work efficiently while managing costs.

Available Models

ModelProviderCostSpeedIntelligenceBest For
AutoFirebuzzLowFastHighGeneral use — let Firebuzz choose the best model
Minimax M2MinimaxLowFastHighBest balance of cost, speed, and quality

Anthropic (Claude)

ModelCostSpeedIntelligenceBest For
Claude Haiku 4.5LowFastMediumQuick tasks and simple edits
Claude Sonnet 4.5HighMediumHighestComplex design work and detailed changes
Claude Opus 4.5HighestMediumHighestMost demanding tasks requiring maximum intelligence

OpenAI (GPT)

ModelCostSpeedIntelligenceBest For
GPT-5.1 Codex MaxMediumMediumHighTechnical changes and code-heavy tasks
GPT-5.2MediumMediumHighReasoning and analytical tasks
GPT-5.1 Codex MiniLowFastMediumFaster and cheaper OpenAI model

Google (Gemini)

ModelCostSpeedIntelligenceBest For
Gemini 2.5 ProMediumMediumHighResearch-heavy tasks with long context
Gemini 3 FlashLowFastestMediumFast and affordable Google model
Gemini 3 Pro PreviewHighFastHighestGoogle's latest preview model with highest intelligence

Other Providers

ModelProviderCostSpeedIntelligenceBest For
Kimi K2 Thinking TurboMoonshotAILowFastHighFast thinking model with reasoning capabilities
Kimi K2 ThinkingMoonshotAILowMediumHighThinking model with reasoning capabilities

How to Switch Models

  1. Look for the model button in the chat input area (shows current model name, e.g., Auto)
  2. Click it to open the model selector
  3. Search by model name or provider
  4. Review cost, speed, and intelligence indicators
  5. Click a model to select it

Each model displays three indicators: Cost (how many credits it uses), Speed (how fast it responds), and Intelligence (how capable it is for complex tasks). More filled dots means higher in that category.

Model Recommendations

Just Getting Started

Use Auto or Minimax M2. These provide excellent results for most tasks while being cost-effective.

Complex Design Work

Use Claude Sonnet 4.5. It excels at understanding design requirements and making sophisticated changes.

Need Speed

Use Gemini 3 Flash. It's the fastest model available, ideal when you're iterating quickly.

Technical Tasks

Use GPT-5.1 Codex Max. It's optimized for code generation and technical modifications.

Understanding the Indicators

Cost

  • Low — Uses fewer credits per message
  • Medium — Moderate credit usage
  • High/Highest — Uses more credits per message

Speed

  • Medium — Standard response time
  • Fast — Quick responses
  • Fastest — Near-instant responses

Intelligence

  • Medium — Good for straightforward tasks
  • High — Capable of complex reasoning
  • Highest — Best for demanding, nuanced tasks

FAQ