Llama 4 Maverick

Mid-tier

by Meta

Llama 4 Maverick is Meta's high-performance open-source model with a 1M context window. It delivers benchmark scores competitive with proprietary mid-tier models while being completely free to download and deploy.

Input Price

Free

per 1M tokens

Output Price

Free

per 1M tokens

Context Window

1M

tokens

Released

2025-04

Open source

Capabilities

textvisioncode

Key Strengths

  • Competitive benchmark scores
  • 1M context window
  • Open source
  • Strong code generation

Best For

  • On-premise deployments
  • Fine-tuning for specific domains
  • Cost-sensitive production
  • Code generation

Benchmark Scores

BenchmarkScoreDescription
MMLU-Pro89.3General knowledge and reasoning across 57 subjects
HumanEval91.7Python code generation and problem solving
GPQA Diamond64.1Graduate-level science questions verified by domain experts
MATH86.7Competition-level mathematics problems
SWE-bench52.8Real-world software engineering tasks from GitHub issues

Scores sourced from public benchmark datasets. See full benchmark leaderboard for all models.

Open Source Model

Llama 4 Maverick is free to download and self-host under the Llama 4 Community License. Hosted API pricing varies by provider (e.g., Together, Fireworks, Groq). See our open source LLM guide for deployment options.

Related Models

View DocumentationCompare ModelsCost CalculatorFull Pricing Guide