How much does JFrog ML (formerly Qwak) cost?
Usage price from
$1.20/minute
- Free plan
- Yes
- Usage price
- $1.20/minute
- Enterprise
- Contact sales
Read from www.qwak.com/old-pricing on Oct 29, 2026. Price changed? Tell usSee the line each price came from
In your favor
- Unlimited usage on Pay As You Go. A flat price with no cap on usage.
- Free plan. Free plan for dev environments: up to 100 QPU/month for a year, unlimited model versions, 1GB in-memory and 10GB storage, chat support. Production use moves to Pay As You Go.
How you buy and get started
Self-serve + salesButtons on their site: Start for Free, Schedule a Demo, Book a Demo, Contact Us
- Sign up
- Self-serve signup via Start for Free "Start for Free"
- Setup
- Not stated
- When sales steps in
- Enterprise plan is custom; contact required "Custom"
- Support
- Chat support on Free "Chat support"
- Support
- Pay As You Go: under 2 hours SLA for critical cases "<2 hours SLA for critical cases"
- Support
- Enterprise: under 1 hour SLA for critical cases "<1 hour SLA for critical cases"
What JFrog ML (formerly Qwak) does
JFrog ML, formerly Qwak and now part of JFrog, combines model management, training, deployment, monitoring, prompt and LLM tooling, and a feature store in one platform. Pricing is based on QPU usage, with a free tier, pay as you go, and a custom enterprise plan.
- Good for
- Data scientists, AI practitioners and ML engineers who want one platform to train, deploy and monitor models, including LLMs, without managing the infrastructure themselves.
- Channels
- Web app, Slack alerts, PagerDuty alerts
- Main product
- JFrog ML platform (MLOps, LLMOps, Feature Store)
- Also includes
- Model registry with unlimited model versions, One-click training and deployment on CPU or GPU, Live API, batch and streaming (Kafka) inference, Automatic inference and training pipelines, Model and feature monitoring, Model A/B testing and traffic splitting
- Connects to
- S3, JFrog Artifactory, HuggingFace, Snowflake, MySQL, Kafka, PostgreSQL, BigQuery, CatBoost, GitHub, OpsGenie, Keras
- Public API
- Not stated
- MCP server
- Not stated
- Experiment tracking
visibility into training parameters and metadata
- Model registry and versioning
Unlimited model versions on JFrog ML model registry
- Pipeline orchestration
Automatic inference and training pipelines
- Model deployment and serving
Deploy models to production at any scale with one click
- Drift and performance monitoring
track input data distribution shifts
- Feature store
The entire feature lifecycle managed in one feature store
- LLM and GenAI support
Manage prompts with a single registry
- Self-hosted or VPC deployment
Self hosted hybrid deployments
The same 8 checks appear on every tool in MLOps platforms. Each tick quotes the vendor's own site.
Security and your data
| Question | Their answer | Where it says so |
|---|---|---|
| Trains AI on your data? | Not stated | |
| HIPAA (signs a BAA) | Not stated | |
| SOC 2 | Not stated | |
| GDPR | Not stated | |
| CCPA | Not stated | |
| PCI | Not stated | |
| ISO 27001 | Not stated | |
| Single sign-on | Enterprise | "SSO login"pricing |
| Trust center | JFrog Trust Center | "[JFrog Trust Center](/trust/)"terms |
Security answers are the vendor's own claims, quoted. We don't audit them.
Every plan
| Plan | Monthly | Billed yearly | Included | Over the allowance | Where it says so |
|---|---|---|---|---|---|
| Free | $0 | – | 100 QPUUp to 100 QPU/month for a year; up to 1GB in-memory and 10GB | – | "Up to 100 QPU/month for a year" |
| Pay As You Go/QPU | $1.20per minute | – | Unlimited | – | "$1.2/QPU" |
| In-memory storageper GB / month | $72per minute | – | – | – | "og ml platform contains two types of storage. super fast in memory storage which is primar" |
| Warehouse storageper TB / month | $60per minute | – | – | – | "orage which is primarily used for online feature store, and warehouse storage, used for of" |
| Enterprise | Quote | – | Unlimited | – | "Custom" |
Free plan: Free plan for dev environments: up to 100 QPU/month for a year, unlimited model versions, 1GB in-memory and 10GB storage, chat support. Production use moves to Pay As You Go.
Fine print
- Billing increment
- Not stated
- Billed beyond usage
- Not stated
- Contract
- Volume discounts for pre-commitments on Enterprise "Volume discounts for pre-commitments"
- Refunds
- Not stated
- Auto renew
- Not stated
- Price change notice
- Not stated
"Not stated" means we read their pricing page and terms and they don't say.
About the company
- Legal name
- JFrog
- Online since
- Jul 8, 2004 Domain registration date
- Customers
- "Chosen by the world's best AI teams" their claim
- Backing
- "Qwak is now officially part of JFrog" their words
JFrog ML (formerly Qwak) alternatives
ClearML$15 /moFree plan · checked Oct 29, 2026
Comet$19 /moFree plan · checked Oct 29, 2026
TrueFoundry$25 /moFree plan · checked Oct 29, 2026
ZenML$39 /moFree plan · checked Oct 29, 2026WEWeights & Biases$60 /moFree plan · checked Oct 29, 2026
Amazon SageMaker$0.04 /minuteFree plan · checked Oct 29, 2026
Azure Machine LearningQuotechecked Oct 29, 2026
BentoMLQuotechecked Oct 29, 2026Reviews elsewhere
We link to reviews instead of copying ratings: those sites don't allow reuse, and we don't test products ourselves.
Questions
Is JFrog ML (formerly Qwak) free?
Yes. Free plan for dev environments: up to 100 QPU/month for a year, unlimited model versions, 1GB in-memory and 10GB storage, chat support. Production use moves to Pay As You Go.