Available Models

datablocks supports multiple high-performance language models optimized for different use cases.

Model Comparison

Model	Context Window	Best For	Speed	Input Cost
qwen	32K tokens	General purpose, coding, multilingual	Fast	$0.60 / 1M
llama	8K tokens	Conversation, instruction following	Very Fast	$0.40 / 1M

Qwen

Our flagship model optimized for a wide range of tasks including coding, multilingual understanding, and complex reasoning.

Strengths

✓Excellent code generation and understanding
✓Strong multilingual capabilities (100+ languages)
✓Large 32K context window for complex documents
✓Advanced reasoning and analysis

Best Use Cases

• Code generation and refactoring
• Technical documentation analysis
• Multilingual content processing
• Complex reasoning tasks
• Long-form content generation

Llama

A fast, efficient model optimized for conversational AI and quick response generation.

Strengths

✓Ultra-fast inference times
✓Excellent instruction following
✓Natural conversational abilities
✓Cost-effective for high-volume use

Best Use Cases

• Chatbots and virtual assistants
• Quick Q&A systems
• Content summarization
• Simple instruction following
• High-volume production workloads

Choosing the Right Model

Choose Qwen if you need:

Advanced reasoning or complex analysis
Code generation or technical documentation
Multilingual support
Large context windows for long documents

Choose Llama if you need:

Fast response times
Simple conversational interactions
High-volume, cost-sensitive workloads
Quick Q&A or summarization tasks

Using Models in API Calls

Specify the model in your API requests using the model parameter:

{
  "model": "qwen",
  "messages": [
    {
      "role": "user",
      "content": "Explain quantum computing"
    }
  ],
  "datablocks": [
    {
      "id": "your-datablock-id",
      "source": "wandb"
    }
  ]
}