Useful capabilities, with meaningful limitations to understand.
Freqtrade is one of the strongest free self-hosted choices for Python users because its release cadence, documentation, backtesting and dry-run workflows are unusually mature. It is operational software rather than a managed strategy: users must design or validate logic, secure local API secrets, patch frequently and avoid exposing the web interface. FreqAI expands modeling tools without establishing profitable outcomes.
Local secret control, public code and explicit UI exposure warnings are positive; users remain responsible for dependencies, host security and exchange scopes.
License, code, maintainers, branches, releases and limitations are public; there is no conventional legal operator disclosure.
A long public history and frequent signed releases support project maturity without proving strategy quality.
Backtesting, dry-run, hyperoptimization, FreqAI, UI, Telegram and spot/futures support create a deep research-to-live workflow.
Docker helps, but Python strategy work and server operation remain technical.
The full project is free, with transparent user-borne infrastructure and exchange costs.
Documentation, Discord and issue workflows are active, without contracted support levels.
The tool provides tests and models, not independently assured strategy returns.
What this score means It evaluates the product and available evidence. It does not predict returns or eliminate trading risk.
Trust profile
Thirteen disclosed factors create the calculated Trust Score.
Public maintainers and the Freqtrade Team are visible; no commercial contracting operator applies to the free project.
Factor score: 62/100A company registration is not structurally required for this community open-source project.
Factor score: 65/100Maintainers and contributors are publicly attributable through the repository.
Factor score: 76/100Repository and release records establish sustained development.
Factor score: 92/100GPL-3.0 use is free.
Factor score: 98/100Code and operational warnings are public; no comprehensive independent audit was located.
Factor score: 74/100Local credential configuration and exchange-specific requirements are documented.
Factor score: 90/100Live trading needs order permissions, not withdrawals; users control the actual exchange scope.
Factor score: 88/100Extensive installation, strategy, exchange, testing and update documentation is public.
Factor score: 98/100No independent evidence establishes profitability for user strategies or FreqAI models.
Factor score: 35/100Community channels and issue tracking exist; response outcomes were not measured.
Factor score: 74/100No qualifying project-wide asset-loss incident was identified; local compromises may not be reported centrally.
Factor score: 55/100The project supplies software; users and exchanges determine jurisdiction-specific trading obligations.
Factor score: 62/100Red flags & concerns
Context that should inform—not replace—your own assessment.
During our latest review. This does not mean the platform is risk-free.
Official Docker documentation says freqUI lacks HTTPS out of the box and recommends an SSH tunnel or VPN.
Official installation guidance says keeping Freqtrade updated is important for exchange API compatibility.
Machine-learning tooling can overfit and does not independently validate a strategy.
What we verified
Checks completed during the latest research cycle.
Freqtrade product and safety research
Company, product, permission and regulatory evidence reviewed separately.
Is Freqtrade legit?
A long-running public repository, signed releases, extensive current docs and visible maintainers establish an active open-source project.
Public maintainers provide project accountability, but there is no conventional company identity or paid service agreement for the core software.
Operator and legal sources: no dedicated source in the current evidence set; see the full source list and stated limitations below.
Is Freqtrade safe?
Security depends on local secret handling, Docker or host patching, exchange-key scopes and keeping freqUI away from the public internet.
How Freqtrade works
Users install through Docker or Python, define strategies, download market data, backtest and dry-run before enabling live trading.
Trading bots and automation features
Freqtrade combines Python strategies, historical data, backtests, hyperoptimization, dry-run, FreqAI and live execution in one self-hosted system.
Supported exchanges
Binance, Bybit, OKX, Kraken, KuCoin, Gate.io, Bitget, Hyperliquid
Pricing
Pricing last checked on 2026-08-19. Check the provider's current pricing before purchase.
There is no software subscription; hosting, data, monitoring and trading fees remain user costs.
Pricing source: no dedicated source in the current evidence set; see the full source list and stated limitations below.
Custody model
Funds remain at connected exchanges while credentials and the runtime remain on user-controlled infrastructure.
Security
Freqtrade provides strong visibility and local control, while explicitly warning that its web interface should not be directly internet-exposed.
Prefer official stable images, bind interfaces locally, use a tunnel or VPN, store secrets outside source control, disable withdrawals and update before exchange API changes.
API and withdrawal permissions
Live trading uses exchange keys stored in the local configuration or environment. Dry-run does not require live-trading keys. Withdrawal status: Not required; configure trade-only exchange permissions.
Security and permission sources: Freqtrade repository (Primary) · Freqtrade releases (Primary) · Freqtrade installation (Primary)
Regulatory information
Freqtrade is software rather than a broker or exchange. Venue eligibility and automated-trading obligations depend on the user's location and account.
Regulatory and service-scope sources: no dedicated source in the current evidence set; see the full source list and stated limitations below.
Performance evidence
The tool provides tests and models, not independently assured strategy returns.
Claim verification
Source, claim type, evidence, status and analyst note are shown separately.
“Free and open source crypto trading bot”
- Company source
- Open original claim
- Checked
- August 19, 2026
- What we found
- The repository is public under GPL-3.0 and contains current releases.
- Analyst conclusion
- License and availability are verified; they do not imply security or profitability.
“Designed to support all major exchanges”
- Company source
- Open original claim
- Checked
- August 19, 2026
- What we found
- Current documentation lists a broad but finite set of supported spot and futures venues.
- Analyst conclusion
- Major is subjective and exchange features differ; use the current compatibility table.
“Strategy optimization by machine learning”
- Company source
- Open original claim
- Checked
- August 19, 2026
- What we found
- Hyperoptimization and FreqAI functionality are documented, but no independent evidence shows improved live returns.
- Analyst conclusion
- Tool existence is verified; performance benefit is not.
Security incident history
Documented events and bounded public-source checks, with entity scope preserved.
Public repository and advisory check
No project-wide credential-loss incident was identified in reviewed primary records.
- Provider response / research note
- Users must still track releases, dependencies and exchange-specific defects affecting their installation.
Strengths & limitations
Strengths
Mature GPL-3.0 project
Strong backtesting and dry-run
Frequent releases
Broad exchange support
Limitations
–Requires Python and operations knowledge
–Local UI exposure risk
–No managed strategy
–No independent return assurance
Who may find it useful
Python strategy developers
Users who want local control
Researchers needing reproducible tests
Who should look elsewhere
–You want provider-managed uptime
–You cannot maintain dependencies
–You expect FreqAI to supply proven returns
Freqtrade alternatives and comparisons
Alternatives are selected for overlapping workflows, with the main distinction shown.
Frequently asked questions
Is Freqtrade open source?+
Yes. It is published under GPL-3.0.
Does Freqtrade keep API keys in a cloud service?+
No. Users run the bot and control its configuration, environment and host.
Can Freqtrade guarantee profitable strategies?+
No. Backtests, dry-run and FreqAI are research tools, not guarantees of live returns.
Sources & evidence
Company-provided information is labeled and not treated as independent validation. Research coverage: 7/9; evidence confidence: Moderate.
- [01]Primary · DocumentationFreqtrade repository
Freqtrade Team · Accessed August 19, 2026 · Last checked 2026-08-19
- [02]Primary · DocumentationFreqtrade releases
Freqtrade Team · Accessed August 19, 2026 · Last checked 2026-08-19
- [03]Primary · DocumentationFreqtrade installation
Freqtrade · Accessed August 19, 2026 · Last checked 2026-08-19
- [04]Primary · Security DocumentationDocker quickstart and UI warning
Freqtrade · Accessed August 19, 2026 · Last checked 2026-08-19
- [05]Primary · DocumentationSupported exchange notes
Freqtrade · Accessed August 19, 2026 · Last checked 2026-08-19
- [06]Primary · DocumentationStrategy customization
Freqtrade Team · Accessed August 19, 2026 · Last checked 2026-08-19
Research changelog
Conclusions may change when new evidence appears.
Reviewed status, repository, license, releases, deployment, exchange support and claims.
Freqtrade: 78/100
Freqtrade is one of the strongest free self-hosted choices for Python users because its release cadence, documentation, backtesting and dry-run workflows are unusually mature. It is operational software rather than a managed strategy: users must design or validate logic, secure local API secrets, patch frequently and avoid exposing the web interface. FreqAI expands modeling tools without establishing profitable outcomes.
Research disclaimer This review is informational and not financial, investment, legal or security advice. Scores evaluate disclosed product qualities and evidence—not profitability, solvency or safety. Features, prices and eligibility can change.
No active affiliate relationships or compensated outbound links are configured. Any future commercial relationship will be disclosed and will not affect scores or conclusions. Read our policy.