Prompt Testing Lab
Test and compare multiple prompts with the same model to find the best prompt for your use case.
Select Model
Select model to test prompts
GPT-4o
Most advanced GPT-4 model with vision and multimodal capabilities
Llama 3.3 70B Instruct
Latest Llama with GPT-4 level performance (Dec 2024)
DeepSeek-V3
Strongest open-source model, beats GPT-4o (Dec 2024)
GPT-o3
Latest reasoning model with advanced problem-solving capabilities
GPT-o4 Mini
Efficient reasoning model for faster processing
GPT-4o Mini
Faster, cheaper version of GPT-4o
GPT Image 1
Advanced image generation model with improved text rendering
GPT-o1
Reinforcement learning trained model for complex reasoning
GPT-o1 Preview
Research preview of o1 series with step-by-step reasoning
ChatGPT-4o Latest
GPT-4o snapshot currently used in ChatGPT
GPT-OSS 120B
Most powerful open-weight model with 117B parameters
GPT-5.2
Latest general-purpose model for complex tasks requiring broad world knowledge
GPT-5.2 Pro
Uses additional compute for harder thinking and better answers
GPT-5.2 Chat Latest
Model powering ChatGPT with conversational optimization
GPT-5 Mini
Balances speed, cost, and capability for cost-optimized applications
GPT-5 Nano
Fastest, cheapest GPT-5 for summarization and classification
Claude Opus 4.5
Premium model combining maximum intelligence with practical performance
Claude Sonnet 4.5
High performance model with advanced reasoning capabilities
Claude Haiku 4.5
Fast model balancing speed with extended thinking capabilities
Claude 3.5 Sonnet (Latest)
Newest Claude 3.5 Sonnet with enhanced computer use capabilities
Claude 3.5 Sonnet (New)
Latest Claude 3.5 Sonnet with improved coding and reasoning
Claude 3 Opus
Most powerful Claude model for complex reasoning tasks
Claude 3.5 Haiku
Fast and efficient Claude model for everyday tasks
Gemini 3 Pro
Complex tasks requiring broad world knowledge and advanced reasoning
Gemini 3 Flash
Pro-level intelligence at Flash speed and pricing
Nano Banana Pro (Gemini 3 Pro Image)
Highest quality image generation model yet
Gemini 2.5 Flash
Best price-performance model with thinking capabilities
Gemini 2.5 Pro
Advanced reasoning with dynamic thinking enabled by default
Gemini 2.5 Flash Native Audio
Native audio model for Live API with complex workflow handling
Gemini 2.0 Flash (Experimental)
Next-generation Gemini with multimodal capabilities
Gemini 1.5 Pro
Advanced reasoning with 2M context window
Gemini 1.5 Flash
Fast multimodal model with 1M context
Qwen 2.5 72B Instruct
Latest from Alibaba with strong multilingual (Sep 2024)
DeepSeek R1
First-generation reasoning model comparable to OpenAI-o1
DeepSeek Coder V2.5
State-of-the-art code completion and infilling model
DeepSeek V3.2
671B parameter MoE model with Multi-head Latent Attention
Mistral Large 2
Latest flagship model with advanced reasoning capabilities
Codestral
Specialized model for code generation and completion
Mistral Nemo
Efficient model for fast inference and deployment
Mistral Large 3
State-of-the-art multimodal model with 41B active parameters
Codestral (August 2025)
Cutting-edge language model for coding with FIM capabilities
Command R+
Advanced model for complex reasoning and generation tasks
Command R
Balanced model for general-purpose tasks
Command R+ (August 2024)
Advanced model for complex reasoning and generation tasks
Command R (August 2024)
Balanced model for general-purpose tasks with RAG optimization
Llama 3.2 90B Vision
Multimodal Llama model with vision capabilities
Llama 3.2 11B Vision
Efficient vision-language model for edge deployment
Grok Beta
xAI's flagship model with real-time information access
Grok Vision Beta
Grok model with advanced image understanding capabilities
Sonar Large 128K Online
Real-time web search with large context window
Sonar Small 128K Online
Fast web search model for quick queries
Sonar Huge 128K Online
Most capable web search model for complex research
GPT-4o (Azure)
GPT-4o via Azure OpenAI Service
GPT-4 Turbo (Azure)
GPT-4 Turbo via Azure OpenAI Service
Claude 3.5 Sonnet (Bedrock)
Claude 3.5 Sonnet via Amazon Bedrock
Amazon Titan Text Premier
Amazon's flagship text generation model
Qwen2.5 72B
Latest Qwen model with enhanced reasoning
Qwen2.5 Coder 32B
Specialized coding model from Alibaba
Llama 3.1 Nemotron 70B
NVIDIA optimized Llama model
ERNIE 4.0 Turbo
Baidu's flagship Chinese language model
GPT-4o
Most advanced GPT-4 model with vision and multimodal capabilities