modelbenchmark.io

Llama-3.1-8B-Instruct (Arkanecloud)

Meta · llama-3-1-8b-instruct

Compare

Compact open Llama model for lightweight chat, drafting, and self-hosting

Specification

most-agreed values

Context
131K
Max output
4K
Released
2024-07-23
Knowledge cutoff
2023-12
Retires
—
Open weights
yes
Input
text
Output
text

Price

US dollars per million tokens · most-agreed

Input
$0.15
Output
$0.15
Cache read
$0.03

Available from 16 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
helicone$0.02$0.05——16K16K—
kilo · meta-llama$0.02$0.04——131K118K—
nano-gpt · meta-llama$0.02$0.04$0.0272—131K16K—
novita$0.02$0.05——16K16K—
novita-ai · meta-llama$0.02$0.05——16K16K—
inference$0.025$0.025——16K4K—
nscale$0.03$0.03—————
openrouter · meta-llama$0.05$0.08$0.025—131K118K—
huggingface · meta-llama$0.06$0.06——131K4K—
ovhcloud · ovhcloud$0.10$0.10——131K131K—
edenai · arkanecloud$0.15$0.15$0.03—131K4K—
perplexity$0.20$0.20——131K131K—
merge-gateway$0.22$0.22——128K2K—
wandb · meta-llama$0.22$0.22$0.22—131K131K—
wandb · wandb$0.22$0.22——131K128K—
oci · oci$0.72$0.72——128K4K—

Price history

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

DateHostField$/M
2026-09-02openrouter · meta-llamaCache read$0.025
2026-10-29openrouter · meta-llamaCache read$0.025
2026-09-02nano-gpt · meta-llamaInput$0.0544
2026-10-19nano-gpt · meta-llamaInput$0.02
2026-09-02openrouter · meta-llamaInput$0.05
2026-10-28openrouter · meta-llamaInput$0.02
2026-10-29openrouter · meta-llamaInput$0.05
2026-09-02nano-gpt · meta-llamaOutput$0.085
2026-10-19nano-gpt · meta-llamaOutput$0.04
2026-09-02openrouter · meta-llamaOutput$0.08
2026-10-28openrouter · meta-llamaOutput$0.04
2026-10-29openrouter · meta-llamaOutput$0.08

More from Meta

most-hosted first