ME

Model profile

Llama 4 Maverick

Built by Meta

Llama 4 Maverick is Meta’s natively multimodal, mixture-of-experts model with 17 billion active parameters, 400 billion total parameters, and 128 experts. It accepts multilingual text and images and produces multilingual text and code.[1]

ReleasedApr 5, 2025[1]
Context window1,000,000 tokens[1]
Directory statusCatalogued
Official model cardVisit source ↗

Developer snapshot

Key specifications

Verified Jul 18, 2026

Active parameters
17B[1]
Total parameters
400B[1]
Experts
128[1]
Training tokens
~22T multimodal[1]
Knowledge cutoff
August 2024[1]
Supported languages
12[1]

Technical profile

What developers should know

01

Architecture and modalities

Maverick is an autoregressive mixture-of-experts model. Meta uses early fusion to combine text and image inputs within the model instead of treating vision as a separate downstream feature.[1]

The published model card lists multilingual text and images as inputs and multilingual text and code as outputs. The 12 supported languages are Arabic, English, French, German, Hindi, Indonesian, Italian, Portuguese, Spanish, Tagalog, Thai, and Vietnamese.[1]

02

Training and data scope

Meta reports approximately 22 trillion pretraining tokens for Maverick. Its disclosed data mix includes publicly available and licensed data plus information from Meta products and services; the pretraining-data cutoff is August 2024.[1]

The one-million-token context window is a model specification. Hosted providers can expose different practical limits, so production integrations should verify the endpoint-specific input, output, image, and rate limits separately.[1]

03

Intended uses and limits

Meta positions the instruction-tuned checkpoint for assistant-style chat and visual reasoning. The model card also names visual recognition, image reasoning, captioning, and general image question answering as intended vision uses.[1]

Meta reports testing image understanding with as many as five input images. Use beyond the listed languages or capabilities requires deployment-specific evaluation and safeguards.[1]

Deployment note

Meta releases Maverick in BF16 and FP8-quantized forms and reports that the FP8 weights fit on one H100 DGX host. Treat that as a vendor deployment reference, not a memory guarantee for every serving stack, batch size, or context length.[2]

Reported evaluation

Selected instruction-tuned results

Vendor-reported evaluations from Meta’s model card. All listed evaluations used BF16 models; scores are not independent llms.help tests.[2]

CategoryBenchmarkEvaluationScore
Image reasoningMMMU0-shot accuracy73.4
CodingLiveCodeBench0-shot pass@143.4
Reasoning & knowledgeMMLU Pro0-shot macro average accuracy80.5
Reasoning & knowledgeGPQA Diamond0-shot accuracy69.8
MultilingualMGSM0-shot average exact match92.3

Before deployment

Llama 4 Community License

Maverick uses Meta’s custom Llama 4 Community License, which includes its own redistribution, attribution, acceptable-use, and additional commercial terms. Review those terms before shipping.[4]

Read the license ↗