Llama 3 70B Tool Use

Use Llama 3 Groq 70B Tool Use model for advanced function calling and tool orchestration tasks. This model achieves 90.76% accuracy on the Berkeley Function Calling Leaderboard, the best performance among open-source 70B models.

What This Skill Does

This skill leverages the Llama 3 Groq 70B Tool Use model to:

When to Use This Skill

Use this skill for:

Instructions

Step 1: Define Available Tools

Structure your tools using the XML format expected by the model:

```xml

<tools>

{

"name": "function_name",

"description": "Clear description of what the function does",

"parameters": {

"type": "object",

"properties": {

"param_name": {

"type": "string",

"description": "Parameter description"

}

},

"required": ["param_name"]

}

}

</tools>

```

Provide complete, accurate function signatures including:

Step 2: Format the System Prompt

Use this exact system prompt structure:

```

You are a function calling AI model. You are provided with function signatures within <tools></tools> XML tags. You may call one or more functions to assist with the user query. Don't make assumptions about what values to plug into functions. For each function call return a json object with function name and arguments within <tool_call></tool_call> XML tags as follows:

<tool_call>

{"name": <function-name>,"arguments": <args-dict>}

</tool_call>

Here are the available tools:

<tools> [INSERT TOOL DEFINITIONS] </tools>

```

Step 3: Configure Sampling Parameters

The model is sensitive to temperature and top_p settings:

**Recommended starting point:**

**For more deterministic outputs:** Lower temperature (0.3-0.4)

**For more creative tool use:** Higher temperature (0.6-0.8)

Adjust based on your use case and test thoroughly.

Step 4: Handle Tool Responses

When the model returns tool calls, execute them and return results in this format:

```xml

<tool_response>

{"id":"call_id","result":{...}}

</tool_response>

```

The model will use these results to continue the conversation or make additional tool calls.

Step 5: Implement Safety Measures

Step 6: Test Edge Cases

Verify the model handles:

Constraints

Example Usage

**User request:** "What's the weather in San Francisco and convert the temperature to Fahrenheit?"

**Model response:**

```xml

<tool_call>

{"name":"get_current_weather","arguments":{"location":"San Francisco, CA","unit":"celsius"}}

</tool_call>

```

**After receiving weather data:**

```xml

<tool_call>

{"name":"convert_temperature","arguments":{"value":22,"from_unit":"celsius","to_unit":"fahrenheit"}}

</tool_call>

```

Access

Notes