Meta Llama 4 Scout (New)

The Llama 4 models leverage a Mixture of Experts (MoE) architecture, enabling efficient and powerful processing capabilities. The meta.llama-4-maverick-17b-128e-instruct-fp8 model is optimized for multimodal understanding, multilingual tasks, coding, tool-calling, and powering agentic systems.

Available in These Regions

Brazil East (Sao Paulo) (dedicated AI cluster only)
UK South (London) (dedicated AI cluster only)
Japan Central (Osaka) (dedicated AI cluster only)
US Midwest (Chicago)

Key Features

Meta Llama 4 Series

Multimodal Capabilities: Llama 4 models are natively multimodal, capable of processing and integrating various data types, including text and images. Input text and images and get a text output.
Multilingual Support: Trained on data encompassing 200 languages, with fine-tuning support for 12 languages including Arabic, English, French, German, Hindi, Indonesian, Italian, Portuguese, Spanish, Tagalog, Thai, and Vietnamese. Image understanding is limited to English.
Efficient Deployment: Small GPU footprint.
Knowledge Cutoff: August 2024
Usage Restrictions: The Llama 4 Acceptable Use Policy restricts their use in the European Union (EU).

Meta Llama 4 Maverick

Architecture: Similar to Meta Llama Scout, this model features 17 billion active parameters but within a larger framework of about 400 billion parameters, using 128 experts.
Context Window: Supports a context length of 512,000 tokens. (Maximum prompt + response length is 512,000 tokens for each run.)
Performance: Matches advanced models in coding and reasoning tasks.

Other Features

On-demand inferencing available in Chicago.
For on-demand inferencing, the response length is capped at 4,000 tokens for each run.
English is the only supported language for the image plus text option.
Multilingual option supported for the text only option.
In the Console, input a .png or .jpg image of 5 MB or less.
Submitting an image without a prompt doesn't work. When you submit an image, you must submit a prompt about that image in the same request. You can then submit follow-up prompts and the model keeps the context of the conversation.
If you host the model in the playground, to add the next image and text, you must clear the chat which results in losing context of the previous conversation by clearing the chat.
For API, input a base64 encoded image in each run. A 512 x 512 image is converted to about 1,610 tokens.

Dedicated AI Cluster for the Model

In the preceding region list, models in regions that aren't marked with (dedicated AI cluster only) have both on-demand and dedicated AI cluster options. For the on-demand option, you don't need clusters and you can reach the model in the Console playground or through the API.

To reach a model through a dedicated AI cluster in any listed region, you must create an endpoint for that model on a dedicated AI cluster. For the cluster unit size that matches this model, see the following table.


Base Model	Fine-Tuning Cluster	Hosting Cluster	Pricing Page Information	Request Cluster Limit Increase
Model Name: Meta Llama 4 Scout OCI Model Name: `meta.llama-4-scout-17b-16e-instruct`	Not available for fine-tuning	Unit Size: Large Generic V2 Required Units: 1	Pricing Page Product Name: Large Meta - Dedicated	Limit Name: `dedicated-unit-llama2-70-count` For Hosting, Request Limit Increase by: 2

Tip

If you don't have enough cluster limits in your tenancy for hosting the Meta Llama 4 Scout model on a dedicated AI cluster, request the limit dedicated-unit-llama2-70-count to increase by 2.
Review the Meta Llama 4 Scout cluster performance benchmarks for different use cases.

Release and Retirement Dates


Model	Release Date	On-Demand Retirement Date	Dedicated Mode Retirement Date
`meta.llama-4-scout-17b-16e-instruct`	2025-05-14	At least one month after the release of the 1^st replacement model.	At least 6 months after the release of the 1^st replacement model.

Important

For a list of all model time lines and retirement details, see Retiring the Models.

Model Parameters

To change the model responses, you can change the values of the following parameters in the playground or the API.

Maximum output tokens

The maximum number of tokens that you want the model to generate for each response. Estimate four characters per token. Because you're prompting a chat model, the response depends on the prompt and each response doesn't necessarily use up the maximum allocated tokens.

Temperature

The level of randomness used to generate the output text.

Tip

Start with the temperature set to 0 or less than one, and increase the temperature as you regenerate the prompts for a more creative output. High temperatures can introduce hallucinations and factually incorrect information.

Top p

A sampling method that controls the cumulative probability of the top tokens to consider for the next token. Assign p a decimal number between 0 and 1 for the probability. For example, enter 0.75 for the top 75 percent to be considered. Set p to 1 to consider all tokens.

Top k

A sampling method in which the model chooses the next token randomly from the top k most likely tokens. A high value for k generates more random output, which makes the output text sound more natural. The default value for k is 0 for Cohere Command models and -1 for Meta Llama models, which means that the model should consider all tokens and not use this method.

Frequency penalty

A penalty that's assigned to a token when that token appears frequently. High penalties encourage fewer repeated tokens and produce a more random output.

For the Meta Llama family models, this penalty can be positive or negative. Positive numbers encourage the model to use new tokens and negative numbers encourage the model to repeat the tokens. Set to 0 to disable.

Presence penalty

A penalty that's assigned to each token when it appears in the output to encourage generating outputs with tokens that haven't been used.

Seed

A parameter that makes a best effort to sample tokens deterministically. When this parameter is assigned a value, the large language model aims to return the same result for repeated requests when you assign the same seed and parameters for the requests.

Allowed values are integers and assigning a large or a small seed value doesn't affect the result. Assigning a number for the seed parameter is similar to tagging the request with a number. The large language model aims to generate the same set of tokens for the same integer in consecutive requests. This feature is especially useful for debugging and testing. The seed parameter has no maximum value for the API, and in the Console, its maximum value is 9999. Leaving the seed value blank in the Console, or null in the API disables this feature.

Warning

The seed parameter might not produce the same result in the long-run, because the model updates in the OCI Generative AI service might invalidate the seed.