OpenPipe

Home

Pricing

Blog

Docs

GitHub

Pricing Plans

Personal

Free

Per-Token Training/Inference

Autoscaling

Metrics + Analytics

500 Training Rows/Dataset

1 Fine-Tuned Model

Get Started

Pro

$50

model/month

Per-Token Training/Inference

Autoscaling

Metrics + Analytics

2k Training Rows/Dataset

Up to 10 Fine-Tuned Models

Get Started

Enterprise

Custom Pricing

Unlimited Training Runs

Dedicated GPUs (No Cold Start)

Metrics + Analytics

20k+ Training Rows/Dataset

Unlimited Fine-Tuned Models

Ongoing Data-Drift Detection

SLA + 24/7 Support

Per-Token Rates

Only pay for the tokens you use.

Base Model

Training

Input

Output

Mistral 7b

$0.0040 / 1K tokens

$0.0012 / 1K tokens

$0.0016 / 1K tokens

Llama 2 7b

$0.0040 / 1K tokens

$0.0012 / 1K tokens

$0.0016 / 1K tokens

Llama 2 13b

$0.0080 / 1K tokens

$0.0024 / 1K tokens

$0.0032 / 1K tokens

GPT-3.5 Turbo*

$0.0080 / 1K tokens

$0.003 / 1K tokens

$0.006 / 1K tokens

*GPT-3.5 Turbo per-token rates are billed directly by OpenAI. OpenPipe provides at-cost access.

Per-Token Rates

Only pay for the tokens you use.

Model

Training

Input

Output

Llama 2 7b

Mistral 7b

$0.0040

/ 1K tokens

$0.0012

/ 1K tokens

$0.0016

/ 1K tokens

$0.0040

/ 1K tokens

$0.0012

/ 1K tokens

$0.0016

/ 1K tokens

Llama 2 13b

$0.0080

/ 1K tokens

$0.0024

/ 1K tokens

$0.0032

/ 1K tokens

GPT-3.5*

$0.0080

/ 1K tokens

$0.003

/ 1K tokens

$0.006

/ 1K tokens