## Specs & Substance

### Spec Sheet

- **Developed by**: Meta  
- **Model family**: Llama  
- **Category**: flagship  
- **Modality**: Text + Vision  
- **Context window**: 1M tokens  
- **Architecture**: MoE FP8  
- **Version**: Maverick  
- **License**: Llama 4 Community License  
- **Pricing**: $0.35 in $1.00 out · $0.17 cache  
- **Released**: Apr 2025  
- **Endpoint**: parasail-llama-4-maverick-instruct-fp8

Meta’s multimodal Llama model with native image understanding.

Llama 4 Maverick is part of the Llama family by Meta, categorized as a **flagship** model. It supports a 1M tokens context window and is built on a MoE FP8 architecture.

Key strengths: reasoning, multimodal. Parasail serves Llama 4 Maverick on a global fleet of current-gen GPUs behind a single OpenAI-compatible endpoint — with per-token pricing, no minimums, and dedicated capacity options when you need guaranteed throughput.

## Integrate

### Drop-in via the OpenAI SDK

Point any OpenAI-compatible chat client at Parasail and change the model name. That's it.

#### Python

```python
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://api.parasail.io/v1"
)

response = client.chat.completions.create(
    model="parasail-llama-4-maverick-instruct-fp8",
    messages=[
        {"role": "user", "content": "Hello, what can you do?"}
    ],
    stream=True,
    stream_options={"include_usage": True},
    top_p=1,
    max_tokens=1000,
    temperature=1
)

for chunk in response:
    if chunk.choices and chunk.choices[0].delta.content is not None:
        print(chunk.choices[0].delta.content, end="", flush=True)
```

#### Bash

```bash
curl https://api.parasail.io/v1/chat/completions \
  -H "Authorization: Bearer $PARASAIL_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "parasail-llama-4-maverick-instruct-fp8",
    "messages": [{"role": "user", "content": "Hello, what can you do?"}],
    "stream": true,
    "max_tokens": 1000
  }'
```

## Explore the Library

- **Model GLM-5.2**  
  Z.AI’s next-gen model for agentic engineering with sustained execution on long-horizon tasks.  
  **$1.40/M in**

- **Model GLM-5.1**  
  Strong reasoning and coding model with a 1M context window at competitive pricing.  
  **$1.40/M in**

- **Model GLM-5**  
  Z.AI’s open model for reasoning and agentic workflows.  
  **$1.00/M in**

## Run Llama 4 Maverick in the Platform

Start with free credits — no credit card. Call any model in minutes.