Compare/MaxHermes vs QwenPaw

AI tool comparison

MaxHermes vs QwenPaw

Which one should you ship with? Here is the side-by-side panel verdict, pricing read, reviewer split, and community vote comparison.

M

AI Assistants

MaxHermes

MiniMax's cloud sandbox AI that builds skills from every task

Mixed

50%

Panel ship

Community

Paid

Entry

MaxHermes is MiniMax's managed cloud deployment of the Hermes agent framework, launched April 16 as what the company calls the world's first "cloud sandbox" AI agent with a built-in learning loop. Powered by M2.7 (a 230B MoE model at $0.30/M tokens), it turns autonomous agent deployment into a zero-config managed service—no API keys to configure, no servers to maintain, no Docker containers to manage. The core innovation is a self-evolving skill library. As MaxHermes completes tasks, it automatically extracts reusable "Skills" saved as structured documents, then self-iterates based on user feedback. Unlike tools with manually predefined capabilities, the skill library dynamically grows. The system also supports persistent cross-session memory, natural-language scheduled tasks, and parallel sub-agent execution for complex workflows. Current integrations target Feishu (Lark), DingTalk, and WeCom—the dominant enterprise messaging platforms in China—making this primarily a Chinese enterprise play for now. But the architectural concept is novel: a cloud-sandboxed agent that owns its own compute environment, memory, and evolving skill set, with no local setup required.

Q

AI Assistants

QwenPaw

Alibaba's open-source personal assistant that runs on your machine across every chat app

Mixed

50%

Panel ship

Community

Paid

Entry

QwenPaw (formerly CoPaw/Tongyi CoPaw) is an open-source personal AI assistant from Alibaba's AgentScope team that rebounded in April 2026 with a v1.1 series of releases and a full ecosystem rebrand. It runs locally on your machine or in the cloud, connects to every major chat platform (DingTalk, Feishu, QQ, Discord, iMessage, and more), and executes scheduled tasks, agentic workflows, and memory-based recall — all from a unified interface. The v1.1.3 and v1.1.4 releases in April brought a backup and restore system, QwenPaw as ACP Server (allowing other agents to call into it), proactive agent messaging, a console plugin system, agent statistics, and a shell evasion guard. The rebrand to QwenPaw signals deeper integration with Alibaba's Qwen model ecosystem, meaning you get native access to Qwen 3 and Qwen 3.5 series models out of the box. The appeal is data sovereignty: everything runs on your infrastructure, conversations stay on your machines, and you configure which channels it monitors. For teams already embedded in Alibaba's cloud stack, this is a natural fit. For everyone else, it's an intriguing open-source alternative to commercial personal assistant platforms — if you're willing to self-host.

Decision
MaxHermes
QwenPaw
Panel verdict
Mixed · 2 ship / 2 skip
Mixed · 2 ship / 2 skip
Community
No community votes yet
No community votes yet
Pricing
$0.30/M tokens (M2.7 model)
Open Source (MIT-compatible)
Best for
MiniMax's cloud sandbox AI that builds skills from every task
Alibaba's open-source personal assistant that runs on your machine across every chat app
Category
AI Assistants
AI Assistants

Reviewer scorecard

Builder
80/100 · ship

The primitive here is clear: a managed agent runtime that auto-extracts reusable Skills from task completions, stored as structured documents — think of it as a self-populating tool registry sitting on top of a 230B MoE model, with no infrastructure tax. The DX bet is that zero-config is worth more than composability, which is the right call for an agentic product aimed at enterprise teams who don't want to babysit Docker containers. The moment of truth is whether the Skill extraction actually generalizes across tasks or just memorizes one-off procedures; that's genuinely novel engineering if it works, and the $0.30/M token pricing is transparent enough that I'm not chasing hidden costs. I'm shipping it cautiously — the integrations are China-enterprise-first (Feishu, DingTalk), so Western teams will find the ecosystem gap real, but the architectural idea of an agent that grows its own capability surface deserves a serious look.

80/100 · ship

The ACP Server capability in v1.1.3 is genuinely interesting — being able to call QwenPaw from other agents creates an orchestration layer you can build on. The multi-channel support is real and well-implemented. If you're in the Alibaba / Qwen ecosystem already, this is a no-brainer deploy.

Skeptic
45/100 · skip

The category is cloud-hosted autonomous agent, and the direct competitors are Zapier's AI agents, Make's AI scenarios, and OpenAI's Assistants with tool use — all of which have broader integration ecosystems on day one. The specific scenario where MaxHermes breaks is any workflow that touches tools outside Feishu, DingTalk, or WeCom, which is the entire Western enterprise market and a large slice of the global one. What kills this in 12 months: MiniMax's own M-series model gets commoditized, the 'self-evolving skill library' turns out to be structured prompt caching with extra marketing, and a better-funded competitor ships the same architecture with Slack and Google Workspace integrations. To earn a ship, MaxHermes needs a publicly verifiable demo showing the skill library generalizing across genuinely distinct task types — not a curated walkthrough.

45/100 · skip

The China-ecosystem platforms (DingTalk, Feishu, QQ) are the primary channels, which narrows the appeal significantly for Western teams. The rebrand from CoPaw to QwenPaw is the third name in two years — signs of product identity confusion. Self-hosting requirements also raise the bar considerably.

Futurist
80/100 · ship

The thesis MaxHermes is betting on: within 2-3 years, enterprise AI value shifts from model capability to accumulated task memory — the agent that has already learned your workflows is worth more than the smarter agent starting fresh. That's a falsifiable, specific bet, and the self-evolving skill library is the technical mechanism for it. The second-order effect, if this works, is that switching costs in enterprise AI compound over time exactly like CRM data lock-in did in the 2000s — the longer you run MaxHermes, the harder it becomes to migrate because your skill library is proprietary. The trend line is the shift from stateless LLM calls to stateful agent infrastructure, and MaxHermes is early on it — the China-first integration set is a constraint today but a strategic beachhead if MiniMax's enterprise market share in APAC grows. The dependency that has to hold: skill extraction has to produce genuinely reusable abstractions, not just logged task histories, which is a hard ML problem they haven't proven publicly.

80/100 · ship

Personal AI assistants that you fully own, run locally, and connect to every communication channel you already use — this is where the market is heading. QwenPaw is one of the most complete implementations of this vision available as open source today.

Founder
45/100 · skip

The buyer here is a Chinese enterprise IT department or a tech-forward ops team running on Feishu or DingTalk — that's a real buyer with a real budget, but it's also a geographically constrained market with a single dominant platform player (ByteDance, which owns Feishu) that could ship competing agent infrastructure at any time. The moat is supposed to be the self-evolving skill library — accumulated workflow knowledge that compounds — but there's no public evidence of a data network effect or proprietary training loop that would make that library defensible against a clone. At $0.30/M tokens the unit economics look fine on paper, but there's no published information on what a typical enterprise workflow costs monthly, which means the pricing page is doing the thing I hate most: making me do math I shouldn't have to do. Ship this when they have three published enterprise case studies, a Slack integration, and a published methodology for how skill extraction actually works under the hood.

No panel take
Creator
No panel take
45/100 · skip

The interface is very developer-facing and the supported channels are enterprise-centric Asian platforms I don't use. The concept is great — a personal assistant you fully own — but the execution doesn't feel polished enough for non-technical creative workflows yet.

Weekly AI Tool Verdicts

Get the next comparison in your inbox

New AI tools ship daily. We compare them before you waste an afternoon.

Bookmarks

Loading bookmarks...

No bookmarks yet

Bookmark tools to save them for later