🐦 Twitter Post Details

Viewing enriched Twitter post

@dair_ai

NEW papers on self-organizing LLM Agents. Assign an agent a role, and it'll follow instructions. Let agents figure out roles themselves, and they'll outperform your design. New research tested this across 25,000 tasks with up to 256 agents. The work shows that self-organizing LLM agents spontaneously develop specialized roles without any predefined hierarchy. A sequential coordination protocol outperformed centralized approaches by 14%, agents generated over 5,000 unique roles organically, and open-source models reached 95% of closed-source quality at significantly lower cost. Most multi-agent frameworks today start by defining roles: planner, coder, reviewer, critic. This paper provides large-scale evidence that the opposite approach works better. Give agents a mission, a protocol, and a capable model. The agents will figure out the rest. Paper: https://t.co/3W2sbJgTH0 Learn to build effective AI agents in our academy: https://t.co/LRnpZN7L4c

Media 1
Media 2
View on Twitter

📊 Media Metadata

{
  "media": [
    {
      "type": "photo",
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2039350842382512455/media_0.png",
      "filename": "media_0.png"
    },
    {
      "type": "photo",
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2039350842382512455/media_1.png",
      "filename": "media_1.png"
    }
  ],
  "processed_at": "2026-04-01T14:46:31.238205",
  "pipeline_version": "2.0"
}

🔧 Raw API Response

{
  "type": "tweet",
  "id": "2039350842382512455",
  "url": "https://x.com/dair_ai/status/2039350842382512455",
  "twitterUrl": "https://twitter.com/dair_ai/status/2039350842382512455",
  "text": "NEW papers on self-organizing LLM Agents.\n\nAssign an agent a role, and it'll follow instructions.\n\nLet agents figure out roles themselves, and they'll outperform your design.\n\nNew research tested this across 25,000 tasks with up to 256 agents.\n\nThe work shows that self-organizing LLM agents spontaneously develop specialized roles without any predefined hierarchy.\n\nA sequential coordination protocol outperformed centralized approaches by 14%, agents generated over 5,000 unique roles organically, and open-source models reached 95% of closed-source quality at significantly lower cost.\n\nMost multi-agent frameworks today start by defining roles: planner, coder, reviewer, critic. This paper provides large-scale evidence that the opposite approach works better. Give agents a mission, a protocol, and a capable model. The agents will figure out the rest.\n\nPaper: https://t.co/3W2sbJgTH0\n\nLearn to build effective AI agents in our academy: https://t.co/LRnpZN7L4c",
  "source": "Twitter for iPhone",
  "retweetCount": 3,
  "replyCount": 1,
  "likeCount": 6,
  "quoteCount": 1,
  "viewCount": 738,
  "createdAt": "Wed Apr 01 14:35:04 +0000 2026",
  "lang": "en",
  "bookmarkCount": 7,
  "isReply": false,
  "inReplyToId": null,
  "conversationId": "2039350842382512455",
  "displayTextRange": [
    0,
    280
  ],
  "inReplyToUserId": null,
  "inReplyToUsername": null,
  "author": {
    "type": "user",
    "userName": "dair_ai",
    "url": "https://x.com/dair_ai",
    "twitterUrl": "https://twitter.com/dair_ai",
    "id": "889050642903293953",
    "name": "DAIR.AI",
    "isVerified": false,
    "isBlueVerified": true,
    "verifiedType": null,
    "profilePicture": "https://pbs.twimg.com/profile_images/1643277398522187778/31dedbLo_normal.jpg",
    "coverPicture": "https://pbs.twimg.com/profile_banners/889050642903293953/1773242460",
    "description": "",
    "location": "",
    "followers": 93891,
    "following": 1,
    "status": "",
    "canDm": true,
    "canMediaTag": true,
    "createdAt": "Sun Jul 23 09:12:45 +0000 2017",
    "entities": {
      "description": {
        "urls": []
      },
      "url": {}
    },
    "fastFollowersCount": 0,
    "favouritesCount": 4299,
    "hasCustomTimelines": true,
    "isTranslator": false,
    "mediaCount": 184,
    "statusesCount": 3058,
    "withheldInCountries": [],
    "affiliatesHighlightedLabel": {},
    "possiblySensitive": false,
    "pinnedTweetIds": [
      "2039350842382512455"
    ],
    "profile_bio": {
      "description": "Democratizing AI research, education, and technologies. New AI learning portal: https://t.co/LRnpZN7L4c",
      "entities": {
        "description": {
          "hashtags": [],
          "symbols": [],
          "urls": [
            {
              "display_url": "academy.dair.ai",
              "expanded_url": "https://academy.dair.ai/",
              "indices": [
                80,
                103
              ],
              "url": "https://t.co/LRnpZN7L4c"
            }
          ],
          "user_mentions": []
        },
        "url": {
          "urls": [
            {
              "display_url": "dair.ai",
              "expanded_url": "https://www.dair.ai/",
              "indices": [
                0,
                23
              ],
              "url": "https://t.co/lkqPZtMU5s"
            }
          ]
        }
      }
    },
    "isAutomated": false,
    "automatedBy": null
  },
  "extendedEntities": {
    "media": [
      {
        "display_url": "pic.twitter.com/vRsJY3fVsx",
        "expanded_url": "https://twitter.com/dair_ai/status/2039350842382512455/photo/1",
        "ext_media_availability": {
          "status": "Available"
        },
        "features": {
          "large": {
            "faces": [
              {
                "h": 66,
                "w": 66,
                "x": 969,
                "y": 732
              }
            ]
          },
          "orig": {
            "faces": [
              {
                "h": 66,
                "w": 66,
                "x": 969,
                "y": 732
              }
            ]
          }
        },
        "id_str": "2039350839207415808",
        "indices": [
          281,
          304
        ],
        "media_key": "3_2039350839207415808",
        "media_results": {
          "id": "QXBpTWVkaWFSZXN1bHRzOgwAAQoAARxNOsiHWlAACgACHE06yUSacUcAAA==",
          "result": {
            "__typename": "ApiMedia",
            "id": "QXBpTWVkaWE6DAABCgABHE06yIdaUAAKAAIcTTrJRJpxRwAA",
            "media_key": "3_2039350839207415808"
          }
        },
        "media_url_https": "https://pbs.twimg.com/media/HE06yIdaUAAu9a6.png",
        "original_info": {
          "focus_rects": [
            {
              "h": 701,
              "w": 1252,
              "x": 0,
              "y": 0
            },
            {
              "h": 1252,
              "w": 1252,
              "x": 0,
              "y": 0
            },
            {
              "h": 1427,
              "w": 1252,
              "x": 0,
              "y": 0
            },
            {
              "h": 1544,
              "w": 772,
              "x": 115,
              "y": 0
            },
            {
              "h": 1544,
              "w": 1252,
              "x": 0,
              "y": 0
            }
          ],
          "height": 1544,
          "width": 1252
        },
        "sizes": {
          "large": {
            "h": 1544,
            "w": 1252
          }
        },
        "type": "photo",
        "url": "https://t.co/vRsJY3fVsx"
      }
    ]
  },
  "card": null,
  "place": {},
  "entities": {
    "hashtags": [],
    "symbols": [],
    "urls": [
      {
        "display_url": "arxiv.org/abs/2603.28990",
        "expanded_url": "https://arxiv.org/abs/2603.28990",
        "indices": [
          866,
          889
        ],
        "url": "https://t.co/3W2sbJgTH0"
      },
      {
        "display_url": "academy.dair.ai",
        "expanded_url": "https://academy.dair.ai/",
        "indices": [
          942,
          965
        ],
        "url": "https://t.co/LRnpZN7L4c"
      }
    ],
    "user_mentions": []
  },
  "quoted_tweet": null,
  "retweeted_tweet": null,
  "isLimitedReply": false,
  "article": null
}