massgen/MassGen

Health v2 · maintenance only

Flags: no_license

Adoption not part of the score

1122 stars · 175 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

MassGen is an open-source multi-agent scaling system that runs in the terminal, orchestrating multiple frontier LLM agents to work on the same task in parallel. Agents observe, critique, and build on each other's work, then vote to select the best collectively validated answer.

Use cases

• orchestrate multiple llm agents to collaborate on a complex task
• improve answer quality through multi-agent consensus and voting
• run autonomous ai agents from the terminal
• scale test-time compute with parallel agent refinement
• coordinate agents that use tools and mcp servers
• compare and validate outputs from frontier models automatically

When to choose

• you want redundant parallel agent attempts with collective validation rather than a single agent run
• you need a terminal-first, pip-installable multi-agent orchestrator supporting multiple LLM backends
• your task benefits from iterative critique and consensus, like research, coding, or analysis

When to avoid

• you need a lightweight single-agent pipeline where multi-agent overhead and token cost are wasteful
• you require a GUI or hosted web interface rather than a terminal tool
• you need strict latency or cost budgets, since parallel agents multiply inference spend

Facets

cli-tool · maturity active

agent-framework llm-inference cli terminal-ui mcp artificial-intelligence large-language-models developer-tools python cli cross-platform multi-agent test-time-scaling llm-orchestration collaborative-ai consensus-voting tool-calling ai-agents command-line

Member repositories

For agents

markdown · JSON · MCP: product_card(name="massgen/MassGen")