modelbenchmark.io

Llama 3.2 3B Instruct (Deep Infra)

Meta · llama-3-2-3b-instruct

Compare

Small instruction-tuned Llama model for multilingual dialogue and agentic retrieval

Specification

most-agreed values

Context
131K
Max output
131K
Released
2024-09-25
Knowledge cutoff
2023-12
Retires
—
Open weights
yes
Input
text
Output
text

Price

US dollars per million tokens · most-agreed

Input
$0.02
Output
$0.02
Cache read
$0.0153

Rate limits

Requests per minute
300

Available from 12 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
deepinfra$0.02$0.02——131K131K—
edenai
$0.02
$0.0509
$0.02
$0.335
——
131K
80K
131K—
inference$0.02$0.02——16K4K—
novita$0.03$0.05——33K32K—
nano-gpt · meta-llama$0.0306$0.0493$0.0153—131K8K—
kilo · meta-llama$0.05$0.33——131K118K—
openrouter · meta-llama$0.05$0.33——131K118K—
cloudflare$0.0509$0.335——80K80K—
cloudflare-workers-ai · cf$0.0509$0.335——80K80K—
together_ai$0.06$0.06——131K131K—
hyperbolic$0.12$0.30——33K33K—
watsonx$0.15$0.15——128K128K—

Price history

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

DateHostField$/M
2026-09-02openrouter · meta-llamaInput$0.05
2026-10-15openrouter · meta-llamaInput$0.0509
2026-10-16openrouter · meta-llamaInput$0.05
2026-10-19openrouter · meta-llamaInput$0.0509
2026-10-25openrouter · meta-llamaInput$0.05
2026-09-02openrouter · meta-llamaOutput$0.33
2026-10-15openrouter · meta-llamaOutput$0.335
2026-10-16openrouter · meta-llamaOutput$0.33
2026-10-19openrouter · meta-llamaOutput$0.335
2026-10-25openrouter · meta-llamaOutput$0.33
2026-10-22edenaiInput$0.02
2026-10-23edenaiInput$0.0509
2026-10-24edenaiInput$0.02
2026-10-30edenaiInput$0.0509
2026-10-31edenaiInput$0.02
2026-10-22edenaiOutput$0.02
2026-10-23edenaiOutput$0.335
2026-10-24edenaiOutput$0.02
2026-10-30edenaiOutput$0.335
2026-10-31edenaiOutput$0.02

More from Meta

most-hosted first