modelbenchmark.io

Llama 4 Maverick 17B 128E Instruct

Meta · llama-4-maverick-17b-128e-instruct-fp8

Compare

Open multimodal Llama model for strong reasoning and fast responses

Specification

most-agreed values

Context
430K
Max output
4K
Released
2025-04-05
Knowledge cutoff
2024-12
Retires
2026-11-19
Open weights
yes
Input
text, image
Output
text

Price

US dollars per million tokens · most-agreed

Input
$0.2687
Output
$0.8263
Cache read
$0.1344

Available from 15 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
lambda_ai$0.05$0.10——131K8K—
abacus · meta-llama$0.14$0.59——1M8K—
deepinfra$0.20$0.80——1M1M—
deepinfra · meta-llama$0.20$0.80——1M16K—
azure$0.25$1.00——1M16K—
azure_ai$0.25$1.00——1M1M2026-11-19
azure-cognitive-services$0.25$1.00——1M16K—
io-net · meta-llama$0.2687$0.8263$0.1344—430K4K—
novita$0.27$0.85——1M8K—
novita-ai · meta-llama$0.27$0.85——1M8K—
watsonx$0.371$1.484——131K8K—
watsonx · meta-llama$0.371$1.484——131K8K—
oci$0.72$0.72——512K16K—
oci · oci$0.72$0.72——1M8K—
meta_llama · meta-llama————1M4K—

Price history

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

DateHostField$/M
2026-09-02io-net · meta-llamaCache read$0.075
2026-10-22io-net · meta-llamaCache read$0.1344
2026-09-02azure_aiInput$1.41
2026-09-17azure_aiInput$0.25
2026-09-02hyperInput$0.274
2026-09-12hyperInput$0.255
2026-09-16hyperInput$0.2519
2026-09-02io-net · meta-llamaInput$0.15
2026-10-22io-net · meta-llamaInput$0.2687
2026-09-02azure_aiOutput$0.35
2026-09-17azure_aiOutput$1.00
2026-09-02hyperOutput$0.8992
2026-09-12hyperOutput$0.8365
2026-09-16hyperOutput$0.8256
2026-09-02io-net · meta-llamaOutput$0.60
2026-10-22io-net · meta-llamaOutput$0.8263
2026-09-11hyperCache read$0.137
2026-09-12hyperCache read$0.1275
2026-09-16hyperCache read$0.1259

Other listings of this model

same model, different key

More from Meta

most-hosted first