Open models ready forproduction workflows
Run inference on any model or compare multiple LLMs side-by-side.
Test performance, evaluate outputs, and choose the perfect model for fine-tuning or production deployment. All models support both inference and fine-tuning workflows.
Inference, comparison, and fine-tuning
Run inference on any model, compare multiple LLMs side-by-side, or fine-tune for your specific needs. All workflows supported.
Run inference instantly
Test any model with real-time inference. No setup required. Get immediate results to evaluate performance and quality before fine-tuning or deployment.
Compare multiple LLMs
Side-by-side comparison of any models. Test the same prompt across different LLMs to find the best fit for your use case, whether for inference or fine-tuning.
Inference or fine-tune
Every model supports both workflows. Run inference to test performance, or fine-tune to customize for your specific needs. Choose your path.
Featured Models
Explore our curated selection of production-ready models. Each model comes with deployment notes, evaluation metrics, and usage guidance.
TinyLlama
TinyLlama 1.1B Chat
Small, fast, and reliable model perfect for testing and learning. Recommended for beginners.
Qwen
Qwen2.5 1.5B Instruct
Small model with excellent Arabic language support. Great for multilingual applications.
Meta
Llama 3.2 3B Instruct
Balanced model with good performance and reasonable resource requirements.
Microsoft
Phi-2
Microsoft Phi-2 (2.7B parameters) - Small, efficient model with excellent performance. MIT license. Great for PPO, QA, chat, and code generation.
Mistral AI
Mistral 7B Instruct v0.3
High-quality instruction model with excellent performance. Requires more GPU memory.
Meta
Llama 3.3 70B
High-performance instruction model
Ready to explore models?Start building with production-ready AI
Run inference on any model, compare multiple LLMs side-by-side, or fine-tune for your specific needs. All models support both inference and fine-tuning workflows.