🐦 Twitter Post Details

Viewing enriched Twitter post

@yifannnwu

Introducing SWE-Together: a multi-turn benchmark built from real user–agent coding sessions. Coding agents are often benchmarked like exam-takers: given the full spec up front, then graded on the final code. But real coding help is a conversation — users clarify goals, add constraints, and correct course along the way. SWE-Together turns real coding work into a reproducible, verifiable benchmark: 109 repo-level tasks curated from 11,260 recorded sessions, replayed with a reactive LLM user simulator that preserves the original user’s intent. We evaluate agents as collaborators, not just patch generators: final pass rate and how many user interventions were needed to get there. In this evaluation snapshot, claude-opus-4.8 currently leads among the 7 agents we tested — achieving the highest pass rate while requiring the fewest user interventions. 📄 Paper: https://t.co/Zp5BSPpLTJ 💻 Code: https://t.co/NPgxCMLdHi 🌐 Website: https://t.co/BK50zRGReE

Media 1
Media 2

📊 Media Metadata

{
  "media": [
    {
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2071976415223050636/media_0.jpg",
      "media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2071976415223050636/media_0.jpg",
      "type": "photo",
      "filename": "media_0.jpg"
    },
    {
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2071976415223050636/media_2.jpg",
      "media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2071976415223050636/media_2.jpg",
      "type": "photo",
      "filename": "media_2.jpg"
    }
  ],
  "processed_at": "2026-07-01T13:01:19.893647",
  "pipeline_version": "2.0"
}

🔧 Raw API Response

{
  "type": "tweet",
  "id": "2071976415223050636",
  "url": "https://x.com/yifannnwu/status/2071976415223050636",
  "twitterUrl": "https://twitter.com/yifannnwu/status/2071976415223050636",
  "text": "Introducing SWE-Together: a multi-turn benchmark built from real user–agent coding sessions.\n\nCoding agents are often benchmarked like exam-takers: given the full spec up front, then graded on the final code. But real coding help is a conversation — users clarify goals, add constraints, and correct course along the way.\n\nSWE-Together turns real coding work into a reproducible, verifiable benchmark: 109 repo-level tasks curated from 11,260 recorded sessions, replayed with a reactive LLM user simulator that preserves the original user’s intent.\n\nWe evaluate agents as collaborators, not just patch generators: final pass rate and how many user interventions were needed to get there.\n\nIn this evaluation snapshot, claude-opus-4.8 currently leads among the 7 agents we tested — achieving the highest pass rate while requiring the fewest user interventions.\n\n📄 Paper: https://t.co/Zp5BSPpLTJ\n💻 Code: https://t.co/NPgxCMLdHi\n🌐 Website: https://t.co/BK50zRGReE",
  "source": "Twitter for iPhone",
  "retweetCount": 27,
  "replyCount": 11,
  "likeCount": 213,
  "quoteCount": 5,
  "viewCount": 20183,
  "createdAt": "Tue Jun 30 15:17:27 +0000 2026",
  "lang": "en",
  "bookmarkCount": 160,
  "isReply": false,
  "inReplyToId": null,
  "conversationId": "2071976415223050636",
  "displayTextRange": [
    0,
    274
  ],
  "inReplyToUserId": null,
  "inReplyToUsername": null,
  "author": {
    "type": "user",
    "userName": "yifannnwu",
    "url": "https://x.com/yifannnwu",
    "twitterUrl": "https://twitter.com/yifannnwu",
    "id": "795044265210560513",
    "name": "Yifan Wu",
    "isVerified": false,
    "isBlueVerified": true,
    "verifiedType": null,
    "profilePicture": "https://pbs.twimg.com/profile_images/1704144330007412736/IRzjjaAF_normal.jpg",
    "coverPicture": "",
    "description": "",
    "location": "",
    "followers": 1096,
    "following": 443,
    "status": "",
    "canDm": true,
    "canMediaTag": true,
    "createdAt": "Sat Nov 05 23:24:58 +0000 2016",
    "entities": {
      "description": {
        "urls": []
      },
      "url": {}
    },
    "fastFollowersCount": 0,
    "favouritesCount": 306,
    "hasCustomTimelines": true,
    "isTranslator": false,
    "mediaCount": 4,
    "statusesCount": 64,
    "withheldInCountries": [],
    "affiliatesHighlightedLabel": {},
    "possiblySensitive": false,
    "pinnedTweetIds": [
      "2071976415223050636"
    ],
    "profile_bio": {
      "description": "吴奕凡; AI Research Scientist @Meta | Ph.D. @penn @picslupenn @GRASPlab.",
      "entities": {
        "description": {
          "user_mentions": [
            {
              "id_str": "",
              "indices": [
                27,
                32
              ],
              "name": "",
              "screen_name": "Meta"
            },
            {
              "id_str": "",
              "indices": [
                41,
                46
              ],
              "name": "",
              "screen_name": "penn"
            },
            {
              "id_str": "",
              "indices": [
                47,
                58
              ],
              "name": "",
              "screen_name": "picslupenn"
            },
            {
              "id_str": "",
              "indices": [
                59,
                68
              ],
              "name": "",
              "screen_name": "GRASPlab"
            }
          ]
        },
        "url": {
          "urls": [
            {
              "display_url": "yifannnwu.com",
              "expanded_url": "http://yifannnwu.com",
              "indices": [
                0,
                23
              ],
              "url": "https://t.co/fNlhStukpY"
            }
          ]
        }
      }
    },
    "isAutomated": false,
    "automatedBy": null
  },
  "extendedEntities": {
    "media": [
      {
        "display_url": "pic.twitter.com/bG7i4z4njV",
        "expanded_url": "https://twitter.com/yifannnwu/status/2071976415223050636/photo/1",
        "ext_master_playlist_only": [],
        "ext_media_availability": {
          "status": "Available"
        },
        "ext_playlists": [],
        "features": {
          "large": {
            "faces": [
              {
                "h": 62,
                "w": 62,
                "x": 721,
                "y": 256
              }
            ]
          },
          "orig": {
            "faces": [
              {
                "h": 62,
                "w": 62,
                "x": 721,
                "y": 256
              }
            ]
          }
        },
        "id_str": "2071976408209870848",
        "indices": [
          275,
          298
        ],
        "media_key": "3_2071976408209870848",
        "media_results": {
          "id": "QXBpTWVkaWFSZXN1bHRzOgwAAQoAARzBI5DCVtAACgACHMEjkmRbcYwAAA==",
          "result": {
            "__typename": "ApiMedia",
            "id": "QXBpTWVkaWE6DAABCgABHMEjkMJW0AAKAAIcwSOSZFtxjAAA",
            "media_key": "3_2071976408209870848"
          }
        },
        "media_url_https": "https://pbs.twimg.com/media/HMEjkMJW0AA4Www.jpg",
        "original_info": {
          "focus_rects": [
            {
              "h": 517,
              "w": 924,
              "x": 0,
              "y": 0
            },
            {
              "h": 924,
              "w": 924,
              "x": 0,
              "y": 0
            },
            {
              "h": 1053,
              "w": 924,
              "x": 0,
              "y": 0
            },
            {
              "h": 1200,
              "w": 600,
              "x": 0,
              "y": 0
            },
            {
              "h": 1200,
              "w": 924,
              "x": 0,
              "y": 0
            }
          ],
          "height": 1200,
          "width": 924
        },
        "sizes": {
          "large": {
            "h": 1200,
            "w": 924
          }
        },
        "type": "photo",
        "url": "https://t.co/bG7i4z4njV"
      }
    ]
  },
  "card": null,
  "place": {},
  "entities": {
    "hashtags": [],
    "symbols": [],
    "urls": [
      {
        "display_url": "arxiv.org/abs/2606.29957",
        "expanded_url": "http://arxiv.org/abs/2606.29957",
        "indices": [
          870,
          893
        ],
        "url": "https://t.co/Zp5BSPpLTJ"
      },
      {
        "display_url": "github.com/Togetherbench/…",
        "expanded_url": "http://github.com/Togetherbench/SWE-Together",
        "indices": [
          902,
          925
        ],
        "url": "https://t.co/NPgxCMLdHi"
      },
      {
        "display_url": "togetherbench.com",
        "expanded_url": "http://togetherbench.com",
        "indices": [
          937,
          960
        ],
        "url": "https://t.co/BK50zRGReE"
      }
    ],
    "user_mentions": []
  },
  "quoted_tweet": null,
  "retweeted_tweet": null,
  "isLimitedReply": false,
  "communityInfo": null,
  "article": null
}