Models / Meta / Llama-3.3-70B-Instruct

Meta · llama

Llama-3.3-70B-Instruct

Popular open Llama workhorse for multilingual chat, coding, and self-hosting

$0.60 in · $1.20 out

per 1M tokens · provider rate closest to the median. No first-party Meta price was listed. The reference price is SambaNova's rate, the provider closest to the median.

Provider input prices run from $0.05 to $1.75 per 1M tokens. Output prices run from $0.23 to $2.75.

Context
128,000
Max output
4,096
Cache read
—
Cache write
—
Reasoning
No
Tool calls
Yes
Structured output
Unknown
Vision
No
Open weights
Yes
License
—
Released
Dec 6, 2024
Updated in source
Dec 6, 2024
Knowledge
2023-12
Providers
27

Hugging Face weights

What this model costs per month

Prices are USD per 1M tokens. A month is 30 days. Cache hits use the cache-read rate when the model has one; otherwise the cache field is ignored.

Estimated monthly cost

Llama-3.3-70B-Instruct

$5.40

Cache hit is ignored because this model has no cache-read price.

monthly = requests/day × 30 × (input tokens × (1 − cache hit) × input price + input tokens × cache hit × cache-read price + output tokens × output price) ÷ 1,000,000

Providers

Base rates from Models.dev. Where a provider lists the model more than once, the row is the listing whose id is closest to the canonical model id.
ProviderInputOutputCache readCache writeStatus
NanoGPT$0.05$0.23$0.025——
Kilo Gateway$0.10$0.32———
DevPass$0.14$0.40———
LLM Gateway$0.14$0.40———
Novita AI$0.14$0.40———
Merge Gateway$0.22$0.50$0.11——
Crusoe$0.25$0.75$0.13——
Cloudflare Workers AI$0.29$2.25———
Neon$0.50$1.50———
Abacus$0.59$0.79———
Hugging Face$0.59$0.79———
OpenRouter$0.59$0.79$0.30——
SambaNovareference$0.60$1.20———
CoreWeave$0.71$0.71$0.71——
Amazon Bedrock$0.72$0.72———
OCI Generative AI$0.72$0.72———
Cortecs$0.72$0.72———
watsonx.ai$0.75$0.75———
Pioneer$0.90$0.90$0.90$0.90—
Prime Inference$1.04$2.25———
evroc$1.15$1.15———
Eden AI$1.23$1.23———
GreenPT$1.25$1.25———
AI/ML API$1.35$1.35———
Tinfoil$1.75$2.75———
Pendra$0$0———
Snowflake Cortex—————

Jobs this model qualifies for