About
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...
Related Models
Pricing
Input
$0.15
per 1M tokens
Output
$0.60
per 1M tokens
Blended
$0.26
per 1M tokens
Cheaper than 64% of models. Median price is $0.54/1M tokens.
Cost Calculator
Tokens per day1M
100K100M
Daily
$0.26
Monthly
$7.87
vs. Similar Models
Qwen: Qwen3 VL 235B A22B InstructQ:0.0
$0.37+41%
GPT-5 mini (minimal)Q:0.0
$0.69+162%
gpt-oss-20B (low)Q:0.0
$0.10-64%
Nova 2.0 Pro Preview (Non-reasoning)Q:+0.1
$3.44+1210%
Performance
100
tokens/sec
Faster than 53% of models
0.67
seconds
Faster than 73% of models
0.67
seconds
Faster than 81% of models
Market Median
94 tok/s
6% faster
Median TTFT
1.11s
40% faster
Throughput/Dollar
383
tok/s per $/1M
Speed Comparison
GPT-5 mini (minimal)
101 tok/s+1%
OpenAI: GPT-5.3-Codex
100 tok/s-1%
GPT-5.1 (Non-reasoning)
101 tok/s+1%
Context Window
1.0M
tokens
Larger than 90% of models
Max Output
16K
tokens
2% of context
Benchmarks
MMLU-Pro
80.9%
GPQA Diamond
67.1%
HLE
4.8%
LiveCodeBench
39.7%
SciCode
33.1%
TerminalBench Hard
6.8%
MATH-500
88.9%
AIME
39.0%
AIME 2025
19.3%
IFBench
43.0%
Long Context Recall
46.0%
Tau2
17.8%
Market AverageTop Score