We use cookies. Cookies and browser storage keep you signed in and remember your preferences. With your permission we also measure how the site is used. You can any time. See our cookie policy.
Meta: Llama 4 Maverick
meta-llama/llama-4-maverickTry Meta: Llama 4 Maverick now
Runs on your balance at the prices above. New accounts start with free usage.
One API. Every model.
Session cost $0.00
Settings
Advanced parameters
Meta: Llama 4 Maverick at a glance
- Input price is 58% lower than the median of 434 Text models.
- Context window of 524,288 tokens, enough for long documents and multi-turn conversations.
- Supports tool (function) calling for agents and integrations.
- Supports structured JSON output with a schema.
- Accepts images as input (vision).
- A typical request (1,000 input / 500 output tokens) costs about $0.00180.
About Meta: Llama 4 Maverick
Meta: Llama 4 Maverick is a multimodal language model from Meta, released on 5 April 2025. It is described as Llama 4 Maverick 17B Instruct (128E), built on a mixture-of-experts architecture with 128 experts and 17 billion active parameters per forward pass. It sits in the text category on api14, and its knowledge cutoff is 31 August 2024.
The model accepts text and image inputs and returns text. Its context window is 1,048,576 tokens, which is large enough to hold very long documents, codebases or multi turn histories in a single request, and it can produce up to 16,384 tokens in one response. Because it handles images as well as text, you can send screenshots, diagrams, scanned pages or photos alongside written instructions and get a single text answer back.
That combination of a very long context and image input points to a few practical patterns. You can feed in lengthy contracts, research papers or log files and ask for summaries, extraction or question answering without chunking first. Screenshots of dashboards or UI mockups can be turned into descriptions or code. Mixed inputs work well for document review, where a page image and its surrounding text both matter. The long output limit also suits tasks that need a full report or a large refactor rather than a short reply.
You call the model through api14's OpenAI compatible API using the model id "meta-llama/llama-4-maverick". The request parameters it supports are frequency_penalty, logit_bias, logprobs, max_tokens, presence_penalty, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_logprobs and top_p. Tools and tool_choice let you wire up function calling, structured_outputs and response_format help you constrain the shape of the reply, and seed plus the sampling controls give you repeatable or tuned generation.
Two limits are worth noting before you commit. Output is capped at 16,384 tokens per response, so work that needs a longer single answer has to be split across calls. The knowledge cutoff is 31 August 2024, so questions about events after that date depend on what you supply in the prompt. If your task is text only and fits comfortably in a smaller context, a lighter model may be simpler to work with.
How to call the Meta: Llama 4 Maverick API
- Create an API key in your dashboard.
- Install the official openai SDK for your language.
- Set baseURL to https://api14.com/api/v1 and model to "meta-llama/llama-4-maverick".
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api14.com/api/v1",
apiKey: process.env.API14_API_KEY,
});
const result = await client.chat.completions.create({
"model": "meta-llama/llama-4-maverick",
"messages": [
{
"role": "user",
"content": "Explain vector databases in two sentences."
}
]
});
console.log(result.choices[0].message.content);
Supported parameters
frequency_penaltylogit_biaslogprobsmax_tokenspresence_penaltyrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_pSpecifications
- Context length
- 524,288
- Max output
- 16,384
- Input
- text, image
- Output
- text
- Knowledge cutoff
- 2024-08-31
Requests on api14 · last 7 days
Meta: Llama 4 Maverick: frequently asked questions
- How much does the Meta: Llama 4 Maverick API cost?
- Meta: Llama 4 Maverick costs Input $0.66 / 1M tokens. It is charged from your USD balance only for successful requests; failed requests are free.
- What is the context window of Meta: Llama 4 Maverick?
- Meta: Llama 4 Maverick reads up to 524,288 tokens per request: prompt, history and documents together. It writes up to 16,384 output tokens per response.
- What can Meta: Llama 4 Maverick do?
- Meta: Llama 4 Maverick accepts text, image and returns text. It supports tool calling, structured JSON output.
- How do I call Meta: Llama 4 Maverick from my code?
- Point any OpenAI SDK at https://api14.com/api/v1 and set the model to "meta-llama/llama-4-maverick". No other code changes. Copy a ready example from the code panel on this page.
- What are alternatives to Meta: Llama 4 Maverick?
- Comparable Text models at a similar price: DeepSeek: DeepSeek V4.1 Flash, DeepSeek: DeepSeek V4 Flash Vision Exp, Google: Gemini 3.1 Flash Lite (batch). Each runs through the same API key, so switching is one string.
Meta: Llama 4 Maverick community
New topicThis community has no topics yet. Open the first one about Meta: Llama 4 Maverick.