modelbenchmark.io

Llama 4 Scout 17B 16E Instruct

Meta · llama-4-scout-17b-16e-instruct

Compare

Open multimodal Llama model for long-context analysis and efficient agents

Specification

most-agreed values

Context
128K
Max output
8K
Released
2025-04-05
Knowledge cutoff
2024-08
Retires
2027-01-31
Open weights
yes
Input
text, image
Output
text

Price

US dollars per million tokens · most-agreed

Input
$0.20
Output
$0.78

Rate limits

Requests per minute
300

Available from 14 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
lambda_ai$0.05$0.10——16K8K—
nscale$0.09$0.29—————
deepinfra$0.10$0.30——328K328K—
deepinfra · meta-llama$0.10$0.30——328K16K—
novita$0.18$0.59——131K131K—
novita-ai · meta-llama$0.18$0.59——131K131K—
together_ai$0.18$0.59——1M1M2026-02-06
azure$0.20$0.78——128K8K—
azure_ai$0.20$0.78——10M16K2027-01-31
azure-cognitive-services$0.20$0.78——128K8K—
cloudflare$0.27$0.85——131K131K—
cloudflare-workers-ai · cf$0.27$0.85——131K16K—
oci$0.72$0.72——192K16K—
oci · oci$0.72$0.72——10.5M8K—

Price history

append-only observations · a listing writes a row only when its price moves

DateHostField$/M
2026-09-02together_aiInput$0.18
2026-09-24together_aiInput$0.18
2026-09-02together_aiOutput$0.59
2026-09-24together_aiOutput$0.59

More from Meta

most-hosted first