Two Open-Source Cloud Agents. One Is an API, One Is a Room
Two Open-Source Cloud Agents. One Is an API, One Is a Room
Source: MiniMax-AI/OpenAgentCore BerriAI/moyai
License: MIT + MIT
Made with AI-assisted production (script, voice, diagrams). No real people depicted. Not affiliated with the project owners. Claims from the project README; demos not re-run by us.
#NoShortAnswers #AIAgents #OpenSource #CodingAgents #Shorts
1 comment · 38 views
把后台编码 agent 部署到自己的服务器上，任务从浏览器或 Slack 下发，改代码、跑测试、开 PR 的全程可以插话修正、断点续跑。
BerriAI/moyai t.co
312 views
Moyai is a self-hosted coding agent for background work from the LiteLLM team. It runs Codex, Claude Code or OpenCode against 100+ model providers. 94 stars.
BerriAI/moyai
6 likes · 401 views
𝗠𝗮𝘀𝘀𝗶𝘃𝗲 𝘀𝗵𝗼𝘂𝘁𝗼𝘂𝘁 𝘁𝗼 Ishaan Jaffer 𝗮𝗻𝗱 𝘁𝗵𝗲 𝘁𝗲𝗮𝗺 𝗮𝘁 𝗕𝗲𝗿𝗿𝗶𝗔𝗜 𝗳𝗼𝗿 𝗼𝗽𝗲𝗻-𝘀𝗼𝘂𝗿𝗰𝗶𝗻𝗴 𝗠𝗼𝘆𝗮𝗶 (BerriAI/moyai)! 🚀
They recently shared a 31-day benchmark showing how they cut their AI coding-agent costs from $101,872 (using proprietary tools) down to ~$21,700—a massive 79% net savings—all while maintaining the exact same output.
𝗪𝗵𝘆 𝗠𝗼𝘆𝗮𝗶 𝗶𝘀 𝗯𝗿𝗶𝗹𝗹𝗶𝗮𝗻𝘁:
1️⃣ Isolated Cloud Workspaces: Runs autonomous coding agents in dedicated sandboxes instead of locked-in setups.
2️⃣ Dynamic Model Routing: Uses @LiteLLM to route simple tasks to fast, low-cost models and heavy tasks to high-tier reasoning models.
3️⃣ Full Cost Ownership: Eliminates third-party markups and puts token-level spending controls in your hands.
This is a massive win for the developer community. But for busy engineering leaders, finding the bandwidth to audit, configure, and integrate new open-source agent infra into existing CI/CD pipelines can be a bottleneck.
𝗧𝗵𝗮𝘁’𝘀 𝗲𝘅𝗮𝗰𝘁𝗹𝘆 𝘄𝗵𝗲𝗿𝗲 𝗼𝘂𝗿 𝗩𝗶𝗯𝗲 𝗖𝗧𝗢 𝘀𝗲𝗿𝘃𝗶𝗰𝗲 𝗰𝗼𝗺𝗲𝘀 𝗶𝗻.
Through Vibe CTO, Formulate AI helps founders, CTOs, and VPs of Engineering set up, integrate, and optimize open-source agent workflows like Moyai—fast, securely, and tailored to your tech stack.
What takes weeks of internal R&D, we handle in days. Plus, our commercial model aligns our success directly with your verified net savings.
📌 𝗪𝗮𝗻𝘁 𝘁𝗼 𝘀𝗲𝗲 𝗵𝗼𝘄 𝗺𝘂𝗰𝗵 𝘆𝗼𝘂𝗿 𝗲𝗻𝗴𝗶𝗻𝗲𝗲𝗿𝗶𝗻𝗴 𝘁𝗲𝗮𝗺 𝗰𝗮𝗻 𝘀𝗮𝘃𝗲? 𝗗𝗠 𝗺𝗲 "𝗔𝗜 𝗖𝗢𝗦𝗧" 𝗼𝗿 𝗱𝗿𝗼𝗽 𝗮 𝗰𝗼𝗺𝗺𝗲𝗻𝘁 𝗯𝗲𝗹𝗼𝘄, 𝗮𝗻𝗱 𝗹𝗲𝘁’𝘀 𝗲𝘅𝗽𝗹𝗼𝗿𝗲 𝗮 𝗾𝘂𝗶𝗰𝗸 𝗽𝗶𝗹𝗼𝘁.
#Engineering #DevOps #LiteLLM #Moyai #OpenSource #FormulateAI #VibeCTO #SoftwareEngineering #AICoding
Here's a demo of Moyai in action, get started here: BerriAI/moyai t.co
1 repost · 109 views
You can get started with Moyai here: BerriAI/moyai
We built Moyai because our Devin bill hit $101,872 in one month
All of it was internal use. Most came from sessions and automations our own engineers started, on models and routing we had no control over t.co
2 likes · 217 views
Mentions
Two Open-Source Cloud Agents. One Is an API, One Is a Room Two Open-Source Cloud Agents. One Is an API, One Is a Room Source: MiniMax-AI/OpenAgentCore BerriAI/moyai License: MIT + MIT Made with AI-assisted production (script, voice, diagrams). No real people depicted. Not affiliated with the project owners. Claims from the project README; demos not re-run by us. #NoShortAnswers #AIAgents #OpenSource #CodingAgents #Shorts
把后台编码 agent 部署到自己的服务器上，任务从浏览器或 Slack 下发，改代码、跑测试、开 PR 的全程可以插话修正、断点续跑。 BerriAI/moyai t.co
Moyai is a self-hosted coding agent for background work from the LiteLLM team. It runs Codex, Claude Code or OpenCode against 100+ model providers. 94 stars. BerriAI/moyai
𝗠𝗮𝘀𝘀𝗶𝘃𝗲 𝘀𝗵𝗼𝘂𝘁𝗼𝘂𝘁 𝘁𝗼 Ishaan Jaffer 𝗮𝗻𝗱 𝘁𝗵𝗲 𝘁𝗲𝗮𝗺 𝗮𝘁 𝗕𝗲𝗿𝗿𝗶𝗔𝗜 𝗳𝗼𝗿 𝗼𝗽𝗲𝗻-𝘀𝗼𝘂𝗿𝗰𝗶𝗻𝗴 𝗠𝗼𝘆𝗮𝗶 (BerriAI/moyai)! 🚀 They recently shared a 31-day benchmark showing how they cut their AI coding-agent costs from $101,872 (using proprietary tools) down to ~$21,700—a massive 79% net savings—all while maintaining the exact same output. 𝗪𝗵𝘆 𝗠𝗼𝘆𝗮𝗶 𝗶𝘀 𝗯𝗿𝗶𝗹𝗹𝗶𝗮𝗻𝘁: 1️⃣ Isolated Cloud Workspaces: Runs autonomous coding agents in dedicated sandboxes instead of locked-in setups. 2️⃣ Dynamic Model Routing: Uses @LiteLLM to route simple tasks to fast, low-cost models and heavy tasks to high-tier reasoning models. 3️⃣ Full Cost Ownership: Eliminates third-party markups and puts token-level spending controls in your hands. This is a massive win for the developer community. But for busy engineering leaders, finding the bandwidth to audit, configure, and integrate new open-source agent infra into existing CI/CD pipelines can be a bottleneck. 𝗧𝗵𝗮𝘁’𝘀 𝗲𝘅𝗮𝗰𝘁𝗹𝘆 𝘄𝗵𝗲𝗿𝗲 𝗼𝘂𝗿 𝗩𝗶𝗯𝗲 𝗖𝗧𝗢 𝘀𝗲𝗿𝘃𝗶𝗰𝗲 𝗰𝗼𝗺𝗲𝘀 𝗶𝗻. Through Vibe CTO, Formulate AI helps founders, CTOs, and VPs of Engineering set up, integrate, and optimize open-source agent workflows like Moyai—fast, securely, and tailored to your tech stack. What takes weeks of internal R&D, we handle in days. Plus, our commercial model aligns our success directly with your verified net savings. 📌 𝗪𝗮𝗻𝘁 𝘁𝗼 𝘀𝗲𝗲 𝗵𝗼𝘄 𝗺𝘂𝗰𝗵 𝘆𝗼𝘂𝗿 𝗲𝗻𝗴𝗶𝗻𝗲𝗲𝗿𝗶𝗻𝗴 𝘁𝗲𝗮𝗺 𝗰𝗮𝗻 𝘀𝗮𝘃𝗲? 𝗗𝗠 𝗺𝗲 "𝗔𝗜 𝗖𝗢𝗦𝗧" 𝗼𝗿 𝗱𝗿𝗼𝗽 𝗮 𝗰𝗼𝗺𝗺𝗲𝗻𝘁 𝗯𝗲𝗹𝗼𝘄, 𝗮𝗻𝗱 𝗹𝗲𝘁’𝘀 𝗲𝘅𝗽𝗹𝗼𝗿𝗲 𝗮 𝗾𝘂𝗶𝗰𝗸 𝗽𝗶𝗹𝗼𝘁. #Engineering #DevOps #LiteLLM #Moyai #OpenSource #FormulateAI #VibeCTO #SoftwareEngineering #AICoding
Here's a demo of Moyai in action, get started here: BerriAI/moyai t.co
You can get started with Moyai here: BerriAI/moyai We built Moyai because our Devin bill hit $101,872 in one month All of it was internal use. Most came from sessions and automations our own engineers started, on models and routing we had no control over t.co