Usage-Based Pricing

Pay for what you use. Credits give you transparent, predictable costs across all AI models and compute resources.

Free

50 credits / mo

250 first month with a work email

Pro

25 credits / $1

+ 250/mo included

Team

28 credits / $1

+ 600/user/mo pooled

LLM Model Pricing

Tokens per credit for each model. Input and output tokens are billed separately. Higher numbers = more tokens per credit.

O OpenAI

Model	Tier	Input Tokens / Credit	Output Tokens / Credit
GPT-5.2 (gpt-5.2) Most advanced reasoning and coding	premium	142.9K	35.7K
GPT-5.1 (gpt-5.1) Enhanced reasoning capabilities	premium	200K	50K
GPT-5 (gpt-5) Frontier model for complex tasks	premium	250K	62.5K
O3 (o3) Advanced reasoning model	premium	250K	62.5K
GPT-5 Mini (gpt-5-mini) Fast and capable for most tasks	standard	1M	250K
GPT-4.1 (gpt-4.1) Reliable workhorse model	standard	250K	62.5K
GPT-4.1 Mini (gpt-4.1-mini) Efficient for routine tasks	standard	1.3M	312.5K
GPT-4o (gpt-4o) Multimodal capabilities	standard	200K	50K
O4 Mini (o4-mini) Efficient reasoning	standard	454.5K	113.6K
GPT-5 Nano (gpt-5-nano) Ultra-fast for simple tasks	economy	5M	1.3M
GPT-4.1 Nano (gpt-4.1-nano) Lightweight and fast	economy	5M	1.3M
GPT-4o Mini (gpt-4o-mini) Fast multimodal processing	economy	3.3M	833.3K

A Anthropic

Model	Tier	Input Tokens / Credit	Output Tokens / Credit
Claude Opus 4.5 (claude-opus-4.5) Most capable Claude model	premium	33.3K	6.7K
Claude Opus 4.1 (claude-opus-4.1) Exceptional reasoning and writing	premium	33.3K	6.7K
Claude Opus 4 (claude-opus-4) Complex analysis and creation	premium	33.3K	6.7K
Claude Sonnet 4.5 (claude-sonnet-4.5) Best balance of speed and capability	standard	166.7K	33.3K
Claude Sonnet 4 (claude-sonnet-4) Excellent for most tasks	standard	166.7K	33.3K
Claude Haiku 4.5 (claude-haiku-4.5) Fast and efficient	economy	625K	125K

G Google

Model	Tier	Input Tokens / Credit	Output Tokens / Credit
Gemini 3 Pro (Preview) (gemini-3-pro-preview) Next-gen multimodal AI	premium	250K	41.7K
Gemini 2.5 Pro (gemini-2.5-pro) Advanced reasoning and coding	premium	400K	50K
Gemini 3 Flash (Preview) (gemini-3-flash-preview) Fast next-gen processing	standard	1M	166.7K
Gemini 2.5 Flash (gemini-2.5-flash) Optimized for speed	standard	3.3M	833.3K
Gemini 2.5 Flash Lite (gemini-2.5-flash-lite) Ultra-efficient processing	economy	6.7M	1.7M
Gemini 2.0 Flash (gemini-2.0-flash) Stable and reliable	economy	5M	1.3M
Gemini 2.0 Flash Lite (gemini-2.0-flash-lite) Lightweight processing	economy	6.7M	1.7M

Training & Compute

Train custom models, deploy dedicated instances, or process batches. All priced in simple credits.

🧠

Training

Train custom AI models on your data

Model Training (GPU)

Train custom models on GPU infrastructure

1 credit = 30 minutes of training

☁️

Deployment

Run dedicated model instances in the cloud

Dedicated Deployment (CPU)

Deploy models on dedicated CPU instances

1 credit = 4 hours of compute

Dedicated Deployment (GPU)

Deploy models on dedicated GPU instances

1 credit = 1 hour of compute

📦

Batch Processing

Process large workloads efficiently

Batch Processing (CPU)

Process large batches on CPU infrastructure

1 credit = 1 hour of compute

Batch Processing (GPU)

Process large batches on GPU infrastructure

1 credit = 15 minutes of compute

⚡

Serverless

Pay only for what you use with our hosted API

Serverless Hosted API

Run inference on our serverless infrastructure

1 credit = 500 seconds of execution

Sandbox Execution

Run Python code safely in isolated Cloud Run containers. Choose a tier based on your workload — from quick transforms to heavy batch processing. Scales to zero when idle.

Lightweight

lightweight

Quick transformations and simple calculations

0.5 credits / 60s

vCPU 0.5

Memory 128 MB

Max timeout 10s

Patch ops 100

Allowed imports

json re math

Standard

standard

Data processing, validation, and moderate computation

1 credit / 60s

vCPU 1

Memory 256 MB

Max timeout 30s

Patch ops 1,000

Allowed imports

json re math datetime collections

Compute

compute

Heavy data processing, statistical analysis, ML inference

2 credits / 60s

vCPU 2

Memory 512 MB

Max timeout 60s

Patch ops 5,000

Allowed imports

json re math datetime collections +3 more

Heavy Compute

heavy

Large-scale data transforms, batch operations

4 credits / 60s

vCPU 4

Memory 1 GB

Max timeout 120s

Patch ops 10,000

Allowed imports

json re math datetime collections +7 more

How Credits Work

1

Get free credits with your plan

Every paid plan includes monthly credits. Pro gets 250 credits/month, Team pools 600 credits/user/month.

2

Use credits for AI features

Code generation, template scanning, and custom model training all consume credits based on usage.

3

Buy more when you need them

Purchase additional credits at your tier's rate. Pro gets 25 credits/$1, Team gets 28 credits/$1.

4

Track usage in real-time

Monitor your credit balance and usage history in your account dashboard.

Ready to get started?

Start free with 50 credits every month — 250 your first month when you sign up with a work email. No credit card required.

Get started free View plans