Back to Models

Llama 4 Scout

metameta/llama-4-scout

How TheRouter serves this differently from the vendor

As the vendor operates it

Meta publishes Llama 4 Scout as open weights under the Llama 4 Community License, with BF16 weights and support for on-the-fly int4 quantization in Meta's release notes.

On TheRouter

TheRouter serves Scout as a managed OpenAI-compatible model route with catalog pricing, context, modalities, and supported parameters coming from standard-models.yaml.

API guide

Chat completion

Standard OpenAI-compatible chat endpoint. Switch only the baseURL and model name β€” 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-scout",
    "messages": [{"role": "user", "content": "Summarize this document."}]
  }'

Vision / image input

Use the standard OpenAI vision message format for image inputs. Meta's Llama 4 card says image understanding was tested with up to 5 input images.

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-scout",
    "messages": [{
      "role": "user",
      "content": [
        {"type": "text", "text": "Extract the key facts from this chart."},
        {"type": "image_url", "image_url": {"url": "https://therouter.ai/assets/vision-sample.png"}}
      ]
    }]
  }'
Fact ledger β€” every claim on this page traces here
sourceURLretrieved
Release dategithub.com/meta-llama β†—2026-08-25verified
Architecturegithub.com/meta-llama β†—2026-08-25verified
Active parametersgithub.com/meta-llama β†—2026-08-25verified
Total parametersgithub.com/meta-llama β†—2026-08-25verified
Context windowgithub.com/meta-llama β†—2026-08-25verified
Input and output modalitiesgithub.com/meta-llama β†—2026-08-25verified
Training token countgithub.com/meta-llama β†—2026-08-25verified
Knowledge cutoffgithub.com/meta-llama β†—2026-08-25verified
Supported languagesgithub.com/meta-llama β†—2026-08-25verified
Licensegithub.com/meta-llama β†—2026-08-25verified
MMLUgithub.com/meta-llama β†—2026-08-25verified
MMLU Progithub.com/meta-llama β†—2026-08-25verified
GPQA Diamondgithub.com/meta-llama β†—2026-08-25verified
MATHgithub.com/meta-llama β†—2026-08-25verified
MMMUgithub.com/meta-llama β†—2026-08-25verified
MathVistagithub.com/meta-llama β†—2026-08-25verified
ChartQAgithub.com/meta-llama β†—2026-08-25verified
MBPPgithub.com/meta-llama β†—2026-08-25verified
Meta releases Llama 4 Maverick and Scout β€” open multimodal MoE herdai.meta.com β†—2026-08-25verified
How is Llama 4 Scout different from Llama 4 Maverick?github.com/meta-llama β†—2026-08-25to verify
Can I self-host Llama 4 Scout?github.com/meta-llama β†—2026-08-25to verify
How many images can Llama 4 Scout handle per request?github.com/meta-llama β†—2026-08-25to verify
Help & contact