Head to head · Assistant proactive · October 2026 research run
Leon vs QwenPaw
QwenPaw scores 62.2 (B) on agent readiness against Leon's 41 (E), and leads in 5 of 7 scored categories. Both do assistant proactive.
Best AI personal assistants that act for one person · All 190 assistants comparisons
Which one, for what
Leon E
Best for A developer who wants a self-hosted assistant with memory, computer use and a local model, and who will read the source.
Also in its favour
- Free to start without a card
- No incidents deducted, where QwenPaw loses 3 points for them
Watch for
No tagged release of the core since 1.0.0-beta.8 on 1 May 2023. The 2.0 Developer Preview is installed by cloning the develop branch
QwenPaw B
Best for One person who wants an assistant on Chinese and Western chat apps with scheduled digests, a consumer mailbox and file-based memory, on their own machine and with a model of their choice.
Ahead on
- Reliability, 70 against 32
- Schema & documentation, 68 against 33
- Agent ergonomics, 67 against 57
- Security & auth, 60 against 34
- Maintenance & community, 73 against 15
Also in its favour
- Runs on your own machine
Watch for
Usage data is on by default, qwenpaw init --defaults accepts it without a prompt, and the opt-out variable is absent from the README and docs
Score by category
| Category | Weight this run | Leon | QwenPaw | Edge |
|---|---|---|---|---|
| Reliability | 16%20 | 32 | 70 | QwenPaw +38 |
| Performance | 10%pending | pending | pending | not scored in this run |
| Schema & documentation | 13%16.2 | 33 | 68 | QwenPaw +35 |
| Agent ergonomics | 13%16.2 | 57 | 67 | QwenPaw +10 |
| Security & auth | 14%17.5 | 34 | 60 | QwenPaw +26 |
| Payments & pricing | 10%12.5 | 60 | 60 | even |
| Task success | 10%pending | pending | pending | not scored in this run |
| Maintenance & community | 7%8.8 | 15 | 73 | QwenPaw +58 |
| Transparency & trust | 7%8.8 | 60 | 56 | Leon +4 |
| Negative events | ≤15 | 0 | -3 | |
| Total | E 41/100 | B 62.2/100 |
Facts side by side
| Fact | Leon | QwenPaw |
|---|---|---|
| Kind | Agent harness | Agent harness |
| Vendor | Louis Grenard (Leon AI) | AgentScope |
| Hosted endpoint | no (local only) | no (local only) |
| Transports | HTTP, stdio | |
| Auth | None | None |
| Pricing | Free | Free |
| x402 | no | no |
| Licence | MIT | Apache-2.0 |
| Read-only variant documented | no | no |
| llms.txt | no | no |
| Last release | 2023-05-01 | 2026-09-11 |
| Terms last updated | no document linked | no document linked |
| Privacy policy last updated | no document linked | no document linked |
| Customer content may train models | ||
| Terms restrict automated access | ||
| Terms restrict benchmarking | ||
| Terms or service can change without notice | ||
| Arbitration or class-action waiver | ||
| Popularity | 18k stars | 36k stars |
Verdicts
Leon
Connected-app credentials are encrypted per profile and the Gmail connection is read-only. The 2.0 Developer Preview has no tagged release, runs tools and a 30-minute proactive pulse with no approval step, listens on every network interface with client authentication off, and sends telemetry by default.
QwenPaw
Mail tools ask for approval by default, automatic mail handling may not delete permanently, and the security docs list what each sandbox backend fails to enforce. Usage data is sent by default, the default approval level passes every tool call no rule flags, and no privacy policy or security advisory is published.
Before you call either
Leon
- Clone the develop branch for 2.0.
npm install --global @leon-ai/cliand docs.getleon.ai install and describe the legacy 1.0 betas - Set
client_interface.auth.enabled: trueand aLEON_PROFILE_TOKENbefore the machine joins a shared network. The server binds 0.0.0.0 - Set
telemetry_enabled: falseandruntime.pulse_enabled: falsein the profileconfig.ymlfor an unattended or private run - Use
availability.tools.allowedto limit tools. Shell, file writes, computer use and browser control are otherwise available to the agent with no approval prompt - Back up
LEON_CONNECTIONS_ENCRYPTION_KEYfrom the profile.env. Without it the stored connections cannot be read
QwenPaw
- Set
QWENPAW_TELEMETRY_DISABLED=truebeforeqwenpaw init --defaultsorqwenpaw app. Both send usage data without asking - Set
approval_leveltoSTRICTorSMARTinagent.json. The defaultAUTOruns every tool call no guard rule flags - Set
dm_policyandgroup_policytoallowlistwithallow_fromon each channel. Both default toopen - Set
QWENPAW_AUTH_ENABLED=truebefore binding beyond 127.0.0.1. Requests from localhost skip login even then - Keep mail tools on
askforsend_message,delete_messageanddelete_thread, and turn on Mail Access Control with automatic handling
Questions
Which is better for AI agents, Leon or QwenPaw?
QwenPaw scores 62.2 (B) on agent readiness against Leon's 41 (E), and leads in 5 of 7 scored categories.
Are Leon and QwenPaw open source?
Yes. Leon is open source (MIT). QwenPaw is open source (Apache-2.0).
Other comparisons with Leon or QwenPaw
- AstrBot vs Leon
- AstrBot vs QwenPaw
- CowAgent vs Leon
- CowAgent vs QwenPaw
- GAIA vs Leon
- GAIA vs QwenPaw
- Gemini Spark vs Leon
- Gemini Spark vs QwenPaw
- Grok Bot vs Leon
- Grok Bot vs QwenPaw
- Haven vs Leon
- Haven vs QwenPaw
- Hermes Agent vs Leon
- Hermes Agent vs QwenPaw
- Instinct vs Leon
- Instinct vs QwenPaw
- IronClaw vs Leon
- IronClaw vs QwenPaw
- Leon vs Lindy
- Leon vs nanobot
- Leon vs NanoClaw
- Leon vs Dots
- Leon vs OpenClaw
- Leon vs PicoClaw
- Leon vs Poke
- Leon vs ZeroClaw
- Leon vs Zo Computer
- Lindy vs QwenPaw
- nanobot vs QwenPaw
- NanoClaw vs QwenPaw
- Dots vs QwenPaw
- OpenClaw vs QwenPaw
- PicoClaw vs QwenPaw
- Poke vs QwenPaw
- QwenPaw vs ZeroClaw
- QwenPaw vs Zo Computer
Machine-readable
- This page as Markdown
/compare/leon-vs-qwenpaw.md· slim.min.md· JSON.json(or sendAccept: text/markdown) - Each listing in full
/api/v1/tools/leon.json·/api/v1/tools/qwenpaw.json - From a terminal
anchor compare leon qwenpaw(the CLI) - Over MCP
compare_tools {"a": "leon", "b": "qwenpaw"}at/mcp, no key