LLMversus
Meta

Llama 4 Maverick

Complete specs, pricing, and benchmark data for Llama 4 Maverick by Meta. Last verified 2026-04-20.

VisionMultimodalJSON ModeFunctionsStreaming
Pricing

Input / 1M tokens

$0.150

Output / 1M tokens

$0.600

Cached Input / 1M

$0.050

Context & Output

Context Window

1.048576M

Max Output

32,768

TTFT

250ms

Speed

90 tok/s

Benchmarks

Arena ELO

1290

Coding ELO

1280

Reasoning ELO

1275

HumanEval

91.5

MMLU

88

MATH

78

GPQA

60.5

Try Llama 4 Maverick API

Start building with Llama 4 Maverick

Get API Access→

Price History (Input $/M tokens)

Strengths
  • +Open-source model with strong performance
  • +Very affordable via hosted providers
  • +1M context window
  • +Mixture-of-experts architecture
Limitations
  • -Requires hosted provider (Together, Fireworks, etc.)
  • -Variable quality across providers
  • -Less refined than proprietary models

Best For

ChatbotsCodingGeneral PurposeOpen Source

Compare Llama 4 Maverick with...

vs Claude Opus 4vs Gemini 2.5 Provs o3vs DeepSeek R1vs o1vs Qwen 3 235B MoEvs Gemini Experimental 1206vs GPT-4.5vs DeepSeek R1 (Groq)vs DeepSeek R1 (Together)vs Grok 3vs Claude Sonnet 4
Official Pricing Page →
Your ad here