Updated 2026๋…„ 4์›”

LlamaIndex Agents for Multi-Agent Systems

LlamaIndex ยท MIT ยท primary language python ยท token-overhead ร—1.4

15-axis capability scores

  • Sequential workflows8/10
  • Parallel workflows6/10
  • Hierarchical workflows7/10
  • Adaptive workflows7/10
  • State management7/10
  • Human-in-the-loop5/10
  • Python support10/10
  • TypeScript support7/10
  • .NET / Java support0/10
  • MCP support7/10
  • A2A support4/10
  • Observability7/10
  • Deployment flexibility7/10
  • Maturity8/10
  • Learning curve (higher = easier)7/10

Tokens per task

LlamaIndex Agents์€(๋Š”) 1.0 ๊ธฐ์ค€์„ (LangGraph) ๋Œ€๋น„ ร—1.4 ํ† ํฐ ์˜ค๋ฒ„ํ—ค๋“œ ๊ณ„์ˆ˜๋ฅผ ๊ฐ€์ง‘๋‹ˆ๋‹ค. ์›” 50,000๊ฐœ ์ž‘์—…, 15,000๊ฐœ ๊ธฐ๋ณธ ํ† ํฐ์˜ ์›Œํฌ๋กœ๋“œ์˜ ๊ฒฝ์šฐ, HITL ๋˜๋Š” ๋ฉ€ํ‹ฐ์—์ด์ „ํŠธ ํŒฌ์•„์›ƒ ์ด์ „์— ์•ฝ ์›” 1050.0M ํ† ํฐ์ž…๋‹ˆ๋‹ค.

์›Œํฌ๋กœ๋“œ์™€ ์„ ํƒํ•œ ๋ชจ๋ธ์— ๋Œ€ํ•œ ๋ณด์ •๋œ ์ถ”์ •์„ ์œ„ํ•ด ๋งˆ๋ฒ•์‚ฌ๋ฅผ ์‹คํ–‰ํ•˜์„ธ์š”.

Run the selector with your workload

Starter scaffold

Buzzi๋Š” LlamaIndex Agents์šฉ 2 ์—์ด์ „ํŠธ hello-world ZIP(Dockerfile, ๊ณ ์ •๋œ ์˜์กด์„ฑ, README, MIT ๋ผ์ด์„ ์Šค)์„ ์ œ๊ณตํ•ฉ๋‹ˆ๋‹ค. ๋งˆ๋ฒ•์‚ฌ๋ฅผ ์™„๋ฃŒํ•  ๋•Œ ์ƒ์„ฑ๋ฉ๋‹ˆ๋‹ค.

Closest alternatives

Ready to commit to LlamaIndex Agents?

Run the wizard, download the scaffold, and book a 30-minute scoping call with Buzzi.ai.

Start the selector