Route order log in sign up
1 pick your primary route

Meta: Llama 4 Scout

meta-llama/llama-4-scout · agent · text+image->text · 327k ctx · added 2025-04-05
$0.1218·0.3645 /M effective hit 0% 1.08B/d

Llama 4 Scout 17B Instruct (16E) is a mixture-of-experts (MoE) language model developed by Meta, activating 17 billion parameters out of a total of 109B. It supports native multimodal input...

view
openrouter/meta-llama/llama-4-scout
Meta: Llama 4 Scout · agent · text+image->text · 327k ctx
1.08B/d 0% $0.1218·0.3645 – 1y
sticker
life 429 tok/d dp age
¢ ¢ᵉ ¢ⁱ ♻
deepinfra/fp8/flex
$0.0800·0.2400 $0.0800·0.2400ⁱ $0.0797·0.2301ᵉ 0% – – <0.1% ✓ – – 1d ·
✓ ★
deepinfra/fp8
$0.1000·0.3000 $0.1000·0.3000ⁱ $0.0997·0.2974ᵉ 0% – – 62% ✓ 472ms 55t/s 3mo ·
⚡
groq
$0.1100·0.3400·0.0550 · $0.1100·0.3400ⁱ $0.1098·0.3376ᵉ · 0% · – – – · ✓ 279ms · 186t/s · ✕2mo ·
deepinfra/fp8/priority
$0.1500·0.4500 $0.1500·0.4500ⁱ $0.1497·0.4476ᵉ 0% – – 38% ✓ 472ms 54t/s 1d ·
novita/bf16
$0.1800·0.5900 · $0.1800·0.5900ⁱ $0.1797·0.5875ᵉ · 0% · – – – · ✓ 694ms · 34t/s · ✕1mo ·
google-vertex/us-east5
$0.2500·0.7000 $0.2500·0.7000ⁱ $0.2498·0.6995ᵉ 0% – – <0.1% ✓ – – 3mo ·
vercel/meta/llama-4-scout
Llama 4 Scout 17B Instruct · agent · text+image->text · 128k ctx
– $0.1000·0.3000 – 1y
hf/meta-llama/Llama-4-Scout-17B-16E-Instruct
Llama-4-Scout-17B-16E-Instruct · agent · text+image->text · 327k ctx
– $0.1400·0.4450 – 1y
compare all models →