Skip to content

Compare model prices

Input, output, cached and batch prices, every route and a month of tokens, side by side.

Llama 3.3 70B Instruct

A month of tokens

  • Llama 3.3 70B Instruct$16.40

100M input and 20M output tokens, list price. Use your own numbers.

Side by side

Llama 3.3 70B Instruct
Input per 1M$0.1
Output per 1M$0.32
Cached inputn/a
Batch in / outn/a
Blended (3:1)$0.155
Context131K
Routes14
ReleasedDec 6, 2024
Leadersboard scoren/a

On the price map

Size: context window, 8K to 2MEqual blended cost (3:1) 243 models, lower left is cheaper

Every route

Llama 3.3 70B Instruct

Way to buyIn / out per 1M
DeepInfra
turbo, fp8
$0.1 / $0.32Same as list
Novita AI
bf16
$0.135 / $0.4+30% vs list
AkashML
fp8 · cached $0.1
$0.2 / $0.52+81% vs list
Parasail
fp8 · cached $0.11
$0.22 / $0.5+87% vs list
SambaNova
Default region
$0.45 / $0.9+263% vs list
Groq
Default region · cached $0.295
$0.59 / $0.79+313% vs list
CoreWeave
fp16 · cached $0.71
$0.71 / $0.71+358% vs list
Google Vertex AI
Default region
$0.72 / $0.72+365% vs list
Google Vertex AI
us-central1
$0.72 / $0.72+365% vs list
Databricks Mosaic AI Gateway + Foundation Model APIs
From the platform pricing page
$0.5 / $1.5+384% vs list
IBM watsonx.ai
From the platform pricing page
$0.753 / $0.753+386% vs list
Cloudflare Workers AI
fp8
$0.293 / $2.25+405% vs list
Snowflake Cortex AI (AI_COMPLETE)
From the platform pricing page
$0.864 / $0.864+457% vs list
Together AI
Default region
$1.04 / $1.04+571% vs list

List prices from providers' pages on the date shown; your contract may differ. Logos via logo.dev; trademarks belong to their owners.

Weekly: LLM API list prices that changed, Thursday mornings.