M

Multica

agent_platform Website ↗ Website ↗ Verified 2026-11-17

📊 Scores

Coding Ability
7.5
Value
9.0
Flexibility
9.0
China Access
7.5

🚀 Quick Start

Install Steps

  1. 1 Sign up at multica.ai for Cloud, or clone the GitHub repo for self-hosting
  2. 2 Run multica setup CLI to connect your local machine
  3. 3 Install the selected coding runtime per current docs; the historical count of 12 commands is not a fixed limit
  4. 4 Create agents with custom instructions and attach skills
  5. 5 Assign issues to agents and watch real-time progress
Register / Get Account ↗
🇨🇳 China Access Note: Self-hostable, but mainland reachability of images/runtimes/models is untested; local execution does not imply sandboxing or zero egress.

💲 Pricing

Plan Price Limits
Self-hosted $0 Additional Multica License conditions apply; server/model/tool costs separate Official →

The Multica License (Apache-2.0 plus conditions) permits free internal organizational use, including whole-team self-hosting; server/model/tool costs are separate. External hosted services, commercial embedding and branding waivers require their respective license/commercial checks. Generic website JSON-LD says $0 without establishing Cloud entitlements; /pricing is a not-found page. Do not infer free Cloud, trials, seats, quotas or an SLA. Verify those amounts and entitlements with the vendor; this table lists only verified $0 self-hosted software. Runs inherit full daemon OS-user permissions, with no filesystem-sandbox guarantee: default Codex danger-full-access and Claude Code bypassPermissions. Windows native Codex sandbox requires explicit opt-in, not a cross-platform guarantee. Cloud stores workspace data; titles/follow-ups send some messages to a vendor-selected third-party LLM. Multica says it does not train AI models on content. Self-hosted servers send a content-free daily usage snapshot by default, disabled with DO_NOT_TRACK=1; model/integration terms still apply. Historical editorial scores retained, not newly measured or re-rated; not a fresh performance or cost benchmark.

Multica

Free software/tier listed; check pricing scope and signup terms

Website ↗

✨ Key Features

Code Completion —
Chat ✅
Agent Mode ✅
Multi-file Edit ✅
MCP Support —
Sub-agent —
Multi-Agent ✅
agent_teams ✅
task_management ✅
persistent_memory —
skill_sharing ✅
runtime_monitoring ✅
open_source —
self_hosted ✅
issue_tracking ✅
realtime_progress ✅
Context Window 取决于所用 runtime
supported_runtimes Claude Code, Codex, Cursor, Copilot, Gemini CLI, Kimi Code, Kiro CLI, Antigravity, OpenCode, OpenClaw, Hermes, Pi
supported_os macOS, Windows, Linux
ide_base Agent 协作平台
source_available ✅

🎯 Best For

Teams needing project management + agent executionSelf-hosted teams able to isolate agent permissions with a dedicated user/container/VMTeams building reusable agent skill libraries

📝 Changelog

2026-11-17 Documentary review: verify free internal self-hosted software, not Cloud pricing from generic $0 schema; disclose privacy egress, telemetry and the Windows sandbox exception
2026-10-13 Version bump v0.3.17 → v0.6.2 (verified via GitHub releases)
2026-05-09 Added to CodePick Agent Collaboration section

📚 Related Guides

🔗 Related Tools

💬 Comments

Last Verified: 2026-11-17 Next Review: 2026-12-17 Confidence: low Report Error →