Meta: Llama 4 Maverick
Analysis Summary
Meta Llama 4 Maverick is Meta's flagship multimodal model, supporting text and image inputs with tool use and function calling across a 1M token context window. Its MMLU-Pro of 0.809 and GPQA of 0.671 reflect strong general reasoning, and its IFBench score of 0.430 and LCR of 0.46 make it one of the better instruction-following models in this batch for content and editorial workflows.
For businesses, Maverick is a strong fit for long-document analysis, multimodal content generation, and structured automation pipelines. Its coding index of 15.6 and agentic index of 12.3 are moderate rather than leading, so it is not the primary choice for heavy software engineering, but it handles a broad range of business tasks competently.
At $0.15 input and $0.60 output per million tokens, it offers strong value for a model with this capability breadth. Teams needing a versatile, cost-effective model for content, research, and tool-use workflows will find Maverick a practical and well-rounded choice.
Assessed June 17, 2026
Editorial notes
Meta Llama 4 Maverick pairs a 1M token context window with vision, tool use, strong instruction-following (IFBench 0.430), and a LiveCodeBench score of 0.397 at $0.15 input per million tokens.
Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?
Performance Profile
How Meta: Llama 4 Maverick compares
Meta: Llama 4 Maverick ranks #177 of 382 AI models we track for overall intelligence, #83 of 111 for coding, #246 of 293 for agentic tasks. Its 1M-token context window is larger than 97% of the models we list. At $0.15 per million input tokens it is cheaper than 57% of comparable models.
About Meta: Llama 4 Maverick
Llama 4 Maverick 17B Instruct (128E) is a high-capacity multimodal language model from Meta, built on a mixture-of-experts (MoE) architecture with 128 experts and 17 billion active parameters per forward..
Capabilities
Performance Indices
Source: Artificial Analysis
Benchmark Scores
Intelligence
Technical
Content
Benchmark data from Artificial Analysis and Hugging Face
How does Meta: Llama 4 Maverick stack up?
Compare side-by-side with other specialist models.
Model Information
| OpenRouter ID |
meta-llama/llama-4-maverick
|
| Provider | meta-llama |
| Model Family | Llama 4 |
| Release Date | April 5, 2025 |
| Context Length | 1,048,576 tokens |
| Max Completion | 16,384 tokens |
| Status | Active |
Pricing
| Token Type | Cost per 1M tokens | Cost per 1K tokens |
|---|---|---|
| Input | $0.15 | $0.000150 |
| Output | $0.60 | $0.000600 |
Live Performance
Live endpoint metrics, refreshed every 30 minutes.
Leaderboard Categories
External Resources
Frequently asked questions about Meta: Llama 4 Maverick
How much does Meta: Llama 4 Maverick cost?
Meta: Llama 4 Maverick costs $0.15 per million input tokens and $0.60 per million output tokens.
What is the context window of Meta: Llama 4 Maverick?
Meta: Llama 4 Maverick has a context window of 1,048,576 tokens (1M).
Is Meta: Llama 4 Maverick good for coding?
On our coding benchmark index, Meta: Llama 4 Maverick ranks #83 of 111 models, placing it in the broader range of the field for code generation and debugging.
What can Meta: Llama 4 Maverick do?
Meta: Llama 4 Maverick supports image/vision input, tool use, and function calling.
Who created Meta: Llama 4 Maverick?
Meta: Llama 4 Maverick is developed by Meta and was released on April 5, 2025.
