Back to Models
Llama 4 Maverick
metameta/llama-4-maverick
How TheRouter serves this differently from the vendor
As the vendor operates it
Meta releases Llama 4 Maverick as open weights for self-hosting and also references it as a Meta AI model. Operators can run the BF16 or FP8 checkpoints directly under the Llama 4 Community License.
On TheRouter
TheRouter serves the model as an OpenAI-compatible chat and vision endpoint at api.therouter.ai/v1. It does not expose Meta's raw checkpoints, self-hosting controls, training-time safety stack, or any Meta AI consumer-product surface.
API guide
Chat completion
Standard OpenAI-compatible chat endpoint. Switch only the baseURL and model name — no SDK changes required.
cURL
curl https://api.therouter.ai/v1/chat/completions \
-H "Authorization: Bearer $THEROUTER_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "meta/llama-4-maverick",
"messages": [{"role": "user", "content": "Hello"}]
}'Vision / image input
Pass up to 5 images per request using the standard OpenAI vision message format. Both URL and base64 data URI inputs are supported.
cURL
curl https://api.therouter.ai/v1/chat/completions \
-H "Authorization: Bearer $THEROUTER_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "meta/llama-4-maverick",
"messages": [{
"role": "user",
"content": [
{"type": "text", "text": "What is in this image?"},
{"type": "image_url", "image_url": {"url": "https://therouter.ai/assets/vision-sample.png"}}
]
}]
}'Fact ledger — every claim on this page traces here
| source | URL | retrieved | |
|---|---|---|---|
| Release date | huggingface.co ↗ | 2026-06-10 | verified |
| Architecture | ai.meta.com ↗ | 2026-06-10 | verified |
| Total parameters | huggingface.co ↗ | 2026-06-10 | verified |
| Active parameters per forward pass | ai.meta.com ↗ | 2026-06-10 | verified |
| Context window | huggingface.co ↗ | 2026-06-10 | verified |
| Training token count | huggingface.co ↗ | 2026-06-10 | verified |
| Knowledge cutoff | huggingface.co ↗ | 2026-06-10 | verified |
| Supported languages | huggingface.co ↗ | 2026-06-10 | verified |
| License | github.com/meta-llama ↗ | 2026-06-10 | verified |
| Multimodal input support | huggingface.co ↗ | 2026-06-10 | verified |
| MMLU | github.com/meta-llama ↗ | 2026-08-09 | verified |
| MMLU Pro | github.com/meta-llama ↗ | 2026-08-09 | verified |
| GPQA Diamond | github.com/meta-llama ↗ | 2026-08-09 | verified |
| MATH | github.com/meta-llama ↗ | 2026-08-09 | verified |
| MMMU | github.com/meta-llama ↗ | 2026-08-09 | verified |
| MathVista | github.com/meta-llama ↗ | 2026-08-09 | verified |
| ChartQA | github.com/meta-llama ↗ | 2026-08-09 | verified |
| MBPP | github.com/meta-llama ↗ | 2026-08-09 | verified |
| Meta releases Llama 4 Maverick and Scout — open multimodal MoE herd | ai.meta.com ↗ | 2026-06-10 | verified |
| What is the difference between Llama 4 Maverick and Llama 4 Scout? | huggingface.co ↗ | 2026-06-10 | to verify |
| Can I self-host Llama 4 Maverick? | github.com/meta-llama ↗ | 2026-06-10 | to verify |
| How many images can I send per API request? | huggingface.co ↗ | 2026-06-10 | to verify |