---
title: "Fireworks AI: High-Performance LLM Inference with Llama and Mixtral in Pixeltable"
date: "2025-07-15"
author: "Pixeltable Team"
tags:
  - Fireworks AI
  - Llama 3.3
  - Mixtral
  - LLM Inference
  - AI Integration
  - Pixeltable
description: "Build production AI applications with Fireworks AI's optimized inference platform. Learn how to use Llama 3.3, Mixtral, and other open-source models with Pixeltable for reliable, scalable LLM workloads."
url: "https://pixeltable.com/blog/fireworks-ai-llm-inference-pixeltable"
---

# Fireworks AI: High-Performance LLM Inference with Llama and Mixtral in Pixeltable

Fireworks AI provides enterprise-grade inference for open-source models with exceptional reliability and competitive pricing. Combined with Pixeltable, you can build production AI applications that scale efficiently.

 
## Fireworks AI: Production-Ready Open-Source Inference

 
**Fireworks AI** has built an inference platform optimized for reliability and performance. Their infrastructure delivers consistent low-latency responses with high availability.

 
 
When combined with Pixeltable's [declarative infrastructure](/blog/unified-multimodal-ai-infrastructure-pixeltable), you get reliable inference with enterprise-grade orchestration.

 
## Available Models

 

 - **Llama 3.3 70B Instruct:** Meta's latest instruction-tuned model

 - **Mixtral 8x7B:** Efficient mixture of experts model

 - **Qwen 2.5:** High-performance multilingual model

 - **FireFunction:** Optimized for function calling

 

 
## Getting Started

 
```bash
# Install required packages
pip install pixeltable fireworks-ai

# Set your Fireworks API key
export FIREWORKS_API_KEY="your-api-key-here"
```

 
## Basic Chat Completions

 
```python
import pixeltable as pxt
from pixeltable.functions.fireworks import chat_completions

pxt.drop_dir('fireworks_demo', force=True)
pxt.create_dir('fireworks_demo')

t = pxt.create_table('fireworks_demo.chat', {'input': pxt.String})

messages = [{'role': 'user', 'content': t.input}]
t.add_computed_column(output=chat_completions(
 messages=messages,
 model='accounts/fireworks/models/llama-v3p3-70b-instruct',
 model_kwargs={'max_tokens': 300, 'temperature': 0.7}
))

t.add_computed_column(response=t.output.choices[0].message.content)

t.insert([{'input': 'Who was President of the US in 1961?'}])
t.select(t.input, t.response).show()
```

 
## Function Calling with FireFunction

 
```python
import pixeltable as pxt
from pixeltable.functions.fireworks import chat_completions

funcs = pxt.create_table('fireworks_demo.functions', {'query': pxt.String})

tools = [{
 "type": "function",
 "function": {
 "name": "get_weather",
 "description": "Get current weather for a location",
 "parameters": {
 "type": "object",
 "properties": {
 "location": {"type": "string", "description": "City name"}
 },
 "required": ["location"]
 }
 }
}]

funcs.add_computed_column(
 response=chat_completions(
 messages=[{'role': 'user', 'content': funcs.query}],
 model='accounts/fireworks/models/firefunction-v2',
 model_kwargs={'tools': tools, 'tool_choice': 'auto'}
 )
)

funcs.add_computed_column(
 function_call=funcs.response.choices[0].message.tool_calls
)

funcs.insert([{'query': "What's the weather like in Tokyo?"}])
funcs.show()
```

 
## Pricing Comparison

 
| Model | Input (per 1M tokens) | Output (per 1M tokens) |
| --- | --- | --- |
| Llama 3.3 70B | $0.90 | $0.90 |
| Mixtral 8x7B | $0.50 | $0.50 |
| FireFunction v2 | $0.90 | $0.90 |

 
## Next Steps

 

 - [Multi-Provider Strategy](/blog/multi-provider-ai-strategy-comparison-guide)

 - [Production RAG Guide](/blog/production-rag-data-centric)

 - [All Integrations](https://docs.pixeltable.com/integrations/frameworks)

 

 
## Resources

 

 - [Pixeltable Fireworks Documentation](https://docs.pixeltable.com/howto/providers/working-with-fireworks)

 - [Fireworks AI Documentation](https://docs.fireworks.ai)

 - [Join our Discord Community](https://discord.com/invite/QPyqFYx2UN)