Maverick
Llama 4Flagship
Tool UseVisionStructured Output
About this model
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...
Performance Tier
Flagship
Maverick is a flagship model from Llama 4 : the most capable in their lineup.
Best-in-class model from this provider. Highest performance across benchmarks, ideal for demanding tasks.
Pricing
This model is included in Elosia plans| Type | per 1M tokens |
|---|---|
| Input (prompt) | $0.150 |
| Output (completion) | $0.600 |
Capabilities
Context Length1.0M
Max Output Tokens16K
TokenizerLlama4
Inputtext, image
Outputtext
Release DateApril 5, 2025
Benchmarks
General Intelligence
MMLU
88.2%
MMLU-Pro
80.5%
GPQA Diamond
69.8%
Mathematics
MATH-500
85%
Programming
HumanEval
89.5%
SWE-bench Verified
55%
Reasoning
IFEval
88.8%
Recommended Use Cases
CodingAnalysisGeneral ChatCreative WritingResearch
Strengths
- Most capable open-weight model from Meta
- Strong general reasoning and coding (HumanEval 89.5%)
- Native multimodal support (text + image)
- Open-weight with commercially permissive license
Limitations
- Lags behind top proprietary models on science reasoning (GPQA 69.8%)
- Large model — requires significant compute for self-hosting
Resources
This model may use your data for training
