🐦 Twitter Post Details

Viewing enriched Twitter post

@jiqizhixin

What if an 8B model could out-reason GPT-5 on math? Introducing PaCoRe. It breaks the sequential bottleneck. Instead of thinking step-by-step, it launches hundreds of parallel reasoning "threads" in each round, compresses their insights, and synthesizes them to guide the next round. The result? It scales compute to millions of tokens without hitting context limits, outperforming frontier systems. The 8B model scored 94.5% on HMMT 2025, beating GPT-5's 93.2%. PaCoRe: Learning to Scale Test-Time Compute with Parallel Coordinated Reasoning StepFun, Tsinghua, Peking Paper: https://t.co/qn8ShtFAwR GitHub: https://t.co/MHanDNRBYB Hugging Face: https://t.co/aN4hgyO7bE Our report: https://t.co/08pf1rxOug 📬 #PapersAccepted by Jiqizhixin

Media 1
Media 2
Media 3
Media 4
Media 5

📊 Media Metadata

{
  "media": [
    {
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2003475431819751675/media_0.jpg?",
      "media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2003475431819751675/media_0.jpg?",
      "type": "photo",
      "filename": "media_0.jpg"
    },
    {
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2003475431819751675/media_1.jpg?",
      "media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2003475431819751675/media_1.jpg?",
      "type": "photo",
      "filename": "media_1.jpg"
    },
    {
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2003475431819751675/media_2.jpg?",
      "media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2003475431819751675/media_2.jpg?",
      "type": "photo",
      "filename": "media_2.jpg"
    },
    {
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2003475431819751675/media_3.jpg?",
      "media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2003475431819751675/media_3.jpg?",
      "type": "photo",
      "filename": "media_3.jpg"
    },
    {
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2003475431819751675/media_4.jpg?",
      "media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2003475431819751675/media_4.jpg?",
      "type": "photo",
      "filename": "media_4.jpg"
    }
  ],
  "processed_at": "2025-12-24T00:49:13.769481",
  "pipeline_version": "2.0"
}

🔧 Raw API Response

{
  "type": "tweet",
  "id": "2003475431819751675",
  "url": "https://x.com/jiqizhixin/status/2003475431819751675",
  "twitterUrl": "https://twitter.com/jiqizhixin/status/2003475431819751675",
  "text": "What if an 8B model could out-reason GPT-5 on math?\n\nIntroducing PaCoRe.\n\nIt breaks the sequential bottleneck. Instead of thinking step-by-step, it launches hundreds of parallel reasoning \"threads\" in each round, compresses their insights, and synthesizes them to guide the next round.\n\nThe result? It scales compute to millions of tokens without hitting context limits, outperforming frontier systems. The 8B model scored 94.5% on HMMT 2025, beating GPT-5's 93.2%.\n\nPaCoRe: Learning to Scale Test-Time Compute with Parallel Coordinated Reasoning\n\nStepFun, Tsinghua, Peking\nPaper: https://t.co/qn8ShtFAwR\nGitHub: https://t.co/MHanDNRBYB\nHugging Face: https://t.co/aN4hgyO7bE\n\nOur report: https://t.co/08pf1rxOug\n\n📬 #PapersAccepted by Jiqizhixin",
  "source": "Twitter for iPhone",
  "retweetCount": 17,
  "replyCount": 3,
  "likeCount": 66,
  "quoteCount": 0,
  "viewCount": 2722,
  "createdAt": "Tue Dec 23 14:39:00 +0000 2025",
  "lang": "en",
  "bookmarkCount": 44,
  "isReply": false,
  "inReplyToId": null,
  "conversationId": "2003475431819751675",
  "displayTextRange": [
    0,
    279
  ],
  "inReplyToUserId": null,
  "inReplyToUsername": null,
  "author": {
    "type": "user",
    "userName": "jiqizhixin",
    "url": "https://x.com/jiqizhixin",
    "twitterUrl": "https://twitter.com/jiqizhixin",
    "id": "819861340294524928",
    "name": "机器之心 JIQIZHIXIN",
    "isVerified": false,
    "isBlueVerified": true,
    "verifiedType": null,
    "profilePicture": "https://pbs.twimg.com/profile_images/895029477192851456/7dh0KKva_normal.jpg",
    "coverPicture": "",
    "description": "",
    "location": "Beijing, China",
    "followers": 13648,
    "following": 793,
    "status": "",
    "canDm": false,
    "canMediaTag": true,
    "createdAt": "Fri Jan 13 10:59:10 +0000 2017",
    "entities": {
      "description": {
        "urls": []
      },
      "url": {}
    },
    "fastFollowersCount": 0,
    "favouritesCount": 4354,
    "hasCustomTimelines": true,
    "isTranslator": false,
    "mediaCount": 3251,
    "statusesCount": 8843,
    "withheldInCountries": [],
    "affiliatesHighlightedLabel": {},
    "possiblySensitive": false,
    "pinnedTweetIds": [
      "1952884324120043869"
    ],
    "profile_bio": {
      "description": "China's leading media & information provider for #AI & #MachineLearning",
      "entities": {
        "description": {
          "hashtags": [
            {
              "indices": [
                49,
                52
              ],
              "text": "AI"
            },
            {
              "indices": [
                55,
                71
              ],
              "text": "MachineLearning"
            }
          ]
        },
        "url": {
          "urls": [
            {
              "display_url": "jiqizhixin.com",
              "expanded_url": "http://www.jiqizhixin.com",
              "indices": [
                0,
                23
              ],
              "url": "https://t.co/Ap70A5wYgg"
            }
          ]
        }
      }
    },
    "isAutomated": false,
    "automatedBy": null
  },
  "extendedEntities": {
    "media": [
      {
        "allow_download_status": {
          "allow_download": true
        },
        "display_url": "pic.twitter.com/5eKpAPhjSt",
        "expanded_url": "https://twitter.com/jiqizhixin/status/2003475431819751675/photo/1",
        "ext_media_availability": {
          "status": "Available"
        },
        "features": {
          "large": {
            "faces": [
              {
                "h": 302,
                "w": 302,
                "x": 485,
                "y": 292
              }
            ]
          },
          "orig": {
            "faces": [
              {
                "h": 302,
                "w": 302,
                "x": 485,
                "y": 292
              }
            ]
          }
        },
        "id_str": "2001210703081017344",
        "indices": [
          280,
          303
        ],
        "media_key": "3_2001210703081017344",
        "media_results": {
          "id": "QXBpTWVkaWFSZXN1bHRzOgwAAQoAARvFuogV2pAACgACG83GSlXWsPsAAA==",
          "result": {
            "__typename": "ApiMedia",
            "id": "QXBpTWVkaWE6DAABCgABG8W6iBXakAAKAAIbzcZKVdaw+wAA",
            "media_key": "3_2001210703081017344"
          }
        },
        "media_url_https": "https://pbs.twimg.com/media/G8W6iBXakAAh4GE.jpg",
        "original_info": {
          "focus_rects": [
            {
              "h": 723,
              "w": 1291,
              "x": 0,
              "y": 0
            },
            {
              "h": 866,
              "w": 866,
              "x": 0,
              "y": 0
            },
            {
              "h": 866,
              "w": 760,
              "x": 0,
              "y": 0
            },
            {
              "h": 866,
              "w": 433,
              "x": 138,
              "y": 0
            },
            {
              "h": 866,
              "w": 1291,
              "x": 0,
              "y": 0
            }
          ],
          "height": 866,
          "width": 1291
        },
        "sizes": {
          "large": {
            "h": 866,
            "w": 1291
          }
        },
        "type": "photo",
        "url": "https://t.co/5eKpAPhjSt"
      }
    ]
  },
  "card": null,
  "place": {},
  "entities": {
    "hashtags": [
      {
        "indices": [
          715,
          730
        ],
        "text": "PapersAccepted"
      }
    ],
    "urls": [
      {
        "display_url": "github.com/stepfun-ai/PaC…",
        "expanded_url": "https://github.com/stepfun-ai/PaCoRe/blob/main/pacore_report.pdf",
        "indices": [
          581,
          604
        ],
        "url": "https://t.co/qn8ShtFAwR"
      },
      {
        "display_url": "github.com/stepfun-ai/PaC…",
        "expanded_url": "https://github.com/stepfun-ai/PaCoRe",
        "indices": [
          613,
          636
        ],
        "url": "https://t.co/MHanDNRBYB"
      },
      {
        "display_url": "huggingface.co/stepfun-ai/PaC…",
        "expanded_url": "https://huggingface.co/stepfun-ai/PaCoRe-8B",
        "indices": [
          651,
          674
        ],
        "url": "https://t.co/aN4hgyO7bE"
      },
      {
        "display_url": "mp.weixin.qq.com/s/40kHtPfrO3OK…",
        "expanded_url": "https://mp.weixin.qq.com/s/40kHtPfrO3OKwfTud2GZaw",
        "indices": [
          688,
          711
        ],
        "url": "https://t.co/08pf1rxOug"
      }
    ]
  },
  "quoted_tweet": null,
  "retweeted_tweet": null,
  "isLimitedReply": false,
  "article": null
}