Hummingbot: Open-Source High-Frequency Trading Bot Framework
Design and deploy automated trading bots on centralized and decentralized exchanges.
- Source repo
- hummingbot/hummingbot
- Stars
- ★ 21k
- Last updated
- 27d ago
- License
- Apache-2.0
- Primary language
- Python
- FA score
- 50/100 · Major gaps
At a glance
- How it runs
- Works with
- Universal · cross-platform
- Cost
- Free, no paid service needed
- Setup effort
- Medium · a few setup steps
- You'll need
- Typical use
- Quant developers: use hbot CLI to quickly deploy and monitor market-making scripts like simple_pmm.
- Not a fit if
- Teams needing fully offline operation: AMM DEX connectors require Gateway middleware
- Users unwilling to manage exchange API keys and wallet keys
- Users wanting zero-setup use without Anaconda or Docker
- Source review
- 50/100 · Major gaps
What does this agent do, and when should you use it?
Hummingbot is an open-source framework for building and running automated trading strategies (bots) on numerous centralized and decentralized exchanges. Users have generated over $34 billion in trading volume across 140+ venues. Licensed under Apache 2.0, the codebase is free and public. It offers multiple strategy frameworks: V2 strategy controllers (e.g., pmm_mister), scripts (e.g., simple_pmm), executors (e.g., position_executor), and legacy V1 strategies (e.g., cross-exchange market making). Connectors standardize interfaces for CLOB CEXs, CLOB DEXs, and AMM DEXs. Deployment options include the hbot CLI, an interactive TUI client, and Docker. It also integrates with Condor, an AI harness for agentic strategies.
Hummingbot provides tools to create and operate trading bots. The hbot CLI lets you create configs, start/stop bots, view status/PnL/logs in a scriptable way. Exchange API keys are stored encrypted via hbot connect. Strategy types include scripts (like scripts/simple_pmm.py) for simple market making, controllers (like controllers/generic/pmm_mister.py) for tunable V2 strategies, executors (like position_executor) that manage order lifecycles, and V1 strategies such as cross_exchange_market_making. Connectors abstract REST/WebSocket APIs across CEXs, DEXs, and AMMs. Docker deployment uses make setup and make deploy to run containers.
- Quant developers: use hbot CLI to quickly deploy and monitor market-making scripts like simple_pmm.
- HFT enthusiasts: run pmm_mister strategy on Binance and tune parameters live.
- DEX arbitrageurs: use Gateway middleware to connect Uniswap, PancakeSwap, etc., for arbitrage.
- Cross-exchange market makers: use V1 cross_exchange_market_making to market make on one venue and hedge on another.
- Strategy researchers: leverage controllers and backtesting to develop and test new strategies.
- AI strategy developers: use Condor to connect LLM decision-making to Hummingbot execution.
How do you install or deploy this agent?
Requires Anaconda or Miniconda. Clone the repo and run 'make install' to create the conda environment and build extensions. Then activate with 'conda activate hummingbot'. Alternatively, use Docker: install Docker Compose, run 'make setup' (choose to include Gateway) and 'make deploy' to start the container.
How do you use this agent?
Use the hbot CLI: create a config with 'hbot create simple_pmm --name conf_paper_bot.yml --set exchange=binance_paper_trade --set trading_pair=BTC-USDT', then start with 'hbot start conf_paper_bot.yml', check with 'hbot status', and stop with 'hbot stop'. To trade live, connect exchange API keys via 'hbot connect binance', then create a strategy controller like 'hbot create pmm_mister --name conf_my_bot.yml --set connector_name=binance --set trading_pair=BTC-USDT --set total_amount_quote=100'.
What are this agent's strengths and limitations?
- Broad connector coverage: CEX, DEX, AMM, supporting 140+ venues.
- Multiple strategy frameworks (scripts, controllers, executors, V1) for flexibility.
- hbot CLI enables non-interactive, scriptable bot operation.
- Docker deployment simplifies environment management.
- Requires Anaconda/Miniconda or Docker, adding setup complexity.
- Live trading needs exchange API keys, with key management risk.
- Some DEX connectors depend on Gateway middleware, extra deployment.
- Documentation is scattered; users may rely on Discord and website.
How does this agent compare with similar options?
Key facts side by side with the most closely related agents.
| Agent | Source review | Form / cost | Stars | Updated | Language | Full support on |
|---|---|---|---|---|---|---|
| Hummingbot: Open-Source High-Frequency Trading Bot Framework This agent | 50 · Major gaps | CLIFree | ★ 21k | 27d ago | Python | — |
| OpenEvolve | 40 · Major gaps | CLIFree + model costs | ★ 7.6k | 5d ago | Python | Claude Code · OpenAI API |
| Cheshire Cat AI | 47 · Major gaps | CLIFree | ★ 3.1k | 2mo ago | Python | — |
| Hindsight | 46 · Major gaps | Self-hosted serviceFreemium | ★ 51k | 2d ago | Python | OpenAI API · Claude API |
How does FollowAgents rate this agent?
Why each dimension lost points
Evidence: README mentions encrypted storage of API keys (keystore password) but does not explain least-privilege principles; user confirmation is present via interactive prompts (e.g., hbot first-use password prompt) but no explicit confirmation for high-risk actions; data flow transparency is partial (anonymous data collection mentioned) but lacks detail; sensitive data handling is adequate (encryption, password prompts) but lacks key management details; dependency security is thin (build dependencies pinned in pyproject.toml) but no vulnerability scanning or audit evidence; external effects are significant (real trading) but no risk warnings or consent mechanisms; rollback is absent; source attribution is clear (Hummingbot Foundation in LICENSE) but publisher unverified. Deductions: lack of least-privilege, user confirmation, data flow transparency details; insufficient dependency security evidence; no risk control for external effects; no rollback mechanism.
Evidence: README and pyproject.toml are consistent on installation and testing, but no runtime consistency guarantees; dependency availability is partial (Anaconda/Miniconda and Docker mentioned) but no lock files or mirrors; failure messages are mentioned (stable exit codes) but no concrete examples. Deductions: insufficient dependency availability evidence; lack of concrete failure message examples.
Evidence: README clearly identifies target audience (algorithmic traders and developers) and scenarios (multiple exchanges, strategy types); capability boundaries are outlined (strategy frameworks, connector types) but not exhaustive; trigger precision is partial (hbot CLI commands and parameters) but lacks detailed trigger conditions; environment fit is good (source, Docker, conda) but OS compatibility not specified. Deductions: trigger precision insufficient; environment fit lacks platform specifics.
Evidence: README has clear structure with quick links, installation, strategies, and connectors; install notes are detailed (source and Docker); naming stability is consistent in README but no naming convention doc; examples and FAQ are provided (simple_pmm example, external FAQ links); known limitations are not listed; license is Apache 2.0 with full text; versioning/changelog is absent; maintenance responsibility is implied via community and governance but not explicit. Deductions: missing known limitations, changelog, and explicit maintenance responsibility.
Evidence: Output usability is good (hbot CLI scriptable, stable exit codes) but no output format examples; marginal value is high (multiple strategies and connectors) but no quantified benefits; cost-benefit is good (open source) but no performance or resource data. Deductions: missing output format examples and performance data.
Evidence: Claims in README (e.g., trading volume, exchange count) lack data sources; cross-source corroboration is partial (links to external docs and community) but no independent verification; fact-inference separation is weak (facts and inferences mixed). Deductions: lack of data sources and independent verification.
- This software is for real trading and carries financial risk; use with caution.
- Sensitive information like API keys is encrypted, but ensure your environment is secure.
- Publisher is unverified; assess trustworthiness independently.
Recent releases
Latest entries from the project's GitHub Releases. Notes are quoted from upstream as written, not rewritten or tested by FollowAgents.
-
Release notes excerpt
What's Changed * Update README exchange tables for v2.16.0 by @fengtality in https://github.com/hummingbot/hummingbot/pull/8402 * Update README for master: exchange tables, Getting Started, Strategies by @fengtality in https://github.com/hummingbot/hummingbot/pull/8403 * Feat/std executor config failures by @cardosofede in https://github.com/hummingbot/hummingbot/pull/8421 * fix / Hyperliquid auth silently accepting wrong private keys (#7866) by @ipezygj in https://github.com/hummingbot/hummingbot/pull/8212 …View on GitHub ↗ -
Release notes excerpt
What's Changed * (feat) Lambdaplex Connector by @petioptrv in https://github.com/hummingbot/hummingbot/pull/7889 * feat(bitget): support LIMIT_MAKER (post-only) orders on both connectors by @isreallee82 in https://github.com/hummingbot/hummingbot/pull/8335 * (fix/perf) binance_perpetual: connector-review correctness & rate-limit fixes by @cardosofede in https://github.com/hummingbot/hummingbot/pull/8291 * fix: correct kucoin_perpetual trade stream WebSocket topic and timestamp field by @ipezygj in https://github.com/hummingbot/hummingbot/pull/8313 …View on GitHub ↗ -
Release notes excerpt
What's Changed * feat: default rate oracle source to gate_io instead of binance by @fengtality in https://github.com/hummingbot/hummingbot/pull/8294 * Feat/lighter connector by @vic-en in https://github.com/hummingbot/hummingbot/pull/8223 * Feat/improve position mode set by @cardosofede in https://github.com/hummingbot/hummingbot/pull/8249 * feat(decibel): fetch trading fees from user_fee_rates API by @WGB5445 in https://github.com/hummingbot/hummingbot/pull/8241 * Fix/hyperliquid spot candles by @cardosofede in https://github.com/hummingbot/hummingbot/pull/8271 …View on GitHub ↗