Llama 4 Models by AI at Meta
2 models2025Up to 10m ctxFrom $0.08/1M input
Details
ResearcherAI at Meta
LicenseLlama 4 Community
Commercial useCommercial use with conditions
Models2
Released2025
Max context10m
Capabilities
VisionAll models
MultimodalAll models
Structured OutputsAll models
Links
WebsiteAbout
Meta's Llama 4 family of large language models, featuring Mixture-of-Experts architectures for efficient inference.
Current Variants
Use-when guidance is derived from seed capabilities, context, release, and replacement fields.
2 in view
Use when the workload needs 1m context, structured outputs, and multimodal inputs.
2025-041m contextstructured outputsmultimodal inputs
Use when the workload needs 10m context, structured outputs, and multimodal inputs.
2025-0410m contextstructured outputsmultimodal inputs
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| Llama 4 Maverick 17B Instruct FP8 | Use when the workload needs 1m context, structured outputs, and multimodal inputs. | 2025-04 | 1m contextstructured outputsmultimodal inputs | Current |
| Llama 4 Scout 17B-16E Instruct | Use when the workload needs 10m context, structured outputs, and multimodal inputs. | 2025-04 | 10m contextstructured outputsmultimodal inputs | Current |
Release Timeline
1 release group2025-04
2 current
Llama 4 Maverick 17B Instruct FP8
Current1m contextstructured outputsmultimodal inputs
Llama 4 Scout 17B-16E Instruct
Current10m contextstructured outputsmultimodal inputs
Specifications(2 models)
| Model | Released | Context | Parameters | Vision | Multimodal | Structured Outputs |
|---|---|---|---|---|---|---|
| Llama 4 Maverick 17B Instruct FP8 | 2025-04 | 1m | 400B (17B active) | Yes | Yes | Yes |
| Llama 4 Scout 17B-16E Instruct | 2025-04 | 10m | 109B (17B active) | Yes | Yes | Yes |
Available From(12 providers)
Pricing
Comparisons
- GPT-4o (08-06) vs Llama 4 Maverick 17B Instruct FP8
- GPT-4o Mini (07-18) vs Llama 4 Scout 17B-16E Instruct
- Claude Sonnet 4.6 vs Llama 4 Maverick 17B Instruct FP8
- Claude 3.5 Haiku vs Llama 4 Scout 17B-16E Instruct
- Gemini 2.5 Flash vs Llama 4 Scout 17B-16E Instruct
- DeepSeek V4 Pro vs Llama 4 Maverick 17B Instruct FP8
- DeepSeek V3 vs Llama 4 Maverick 17B Instruct FP8
- Llama 4 Maverick 17B Instruct FP8 vs Grok 4
Frequently Asked Questions
- What is Llama 4 used for?
- Llama 4 is used for vision and multimodal work and structured outputs. The family description and listed model capabilities point to those workloads as the best fit.
- How does Llama 4 compare to Chameleon?
- Llama 4 by AI at Meta is strongest where you need vision and multimodal work, while Chameleon by AI at Meta is the closest related family to check for coding. Llama 4 has 2 listed variants and reaches up to 10m context, while Chameleon reaches up to 4k context, so compare the specs and pricing tables before choosing a production model.
- Which Llama 4 model should I use?
- For the lowest listed input price, start with Llama 4 Scout 17B-16E Instruct through DeepInfra at $0.08/1M input tokens. For the most capable/latest local choice, evaluate Llama 4 Scout 17B-16E Instruct with 10m context and structured outputs and multimodal inputs.
