Meta
Llama 4 Maverick
Cost-sensitive open-weight Llama 4 variant aimed at multimodal general-purpose deployments.
Our editorial assessment, based on the public sources listed below. Last reviewed .
- Overall rank
- #76
- Tier
- B
- Score
- 65.7
- Main category
- Overall
Strengths
- Open weights for self-hosting under Meta's Llama 4 Community License
- Multimodal coverage out of the box
- Lower hosting cost than frontier closed APIs
Considerations
- Benchmark scores trail Scout and top-tier frontier models
- Requires GPU infrastructure to self-host effectively
Price
Open weights (Llama 4 Community License); API $0.19 input / $0.65 output per 1M tokens via OpenRouter
Shown for reference — price is not part of the score.
Score breakdown
- Performance
- 68.0
- 55% of score
- Reputation
- 68.0
- 35% of score
- Recency
- 45.0
- 10% of score
The overall score combines these sub-scores using the weights shown. How we score →
Verification status
Each field carries its own re-verification date. Amber means older than 90 days — treat those figures as stale until re-checked.
- Pricing
- Benchmark
- Reputation
- Feature list
Checked against the sources listed below.
Sources
- Codersera: Llama 4 Guide 2026 (Scout, Maverick, Behemoth, Muse Spark) (opens in a new tab)Accessed:
- Llama (language model) - Wikipedia (opens in a new tab)Accessed:
- Meta Llama (opens in a new tab)Accessed:
- DemandSphere (opens in a new tab)Accessed:
- 22 AI Frontier Models Compared for 2026 - TeamAI (opens in a new tab)Accessed:
Show 1 more source
Similar models
- Qwen3.5-397B-A17BAlibaba#72 · Tier B · Score 68.5Compare
- Ministral 3 14BMistral AI#74 · Tier B · Score 68.3Compare
- Ministral 3 8BMistral AI#78 · Tier B · Score 64.3Compare
- Llama 4 ScoutMeta#80 · Tier B · Score 59.5Compare