🐦 Twitter Post Details

Viewing enriched Twitter post

@wbx_life

🔥 The era of ultra-efficient reasoning is here: our Nemotron-Cascade-2-30B-A3B currently trending at #1 on huggingface 🤗 🥇 Gold Medal-level performance on IMO 2025, IOI 2025, and ICPC World Finals 2025 -- all from a model with only 3B active parameters. 🤯 ⚡ SOTA alignment and instruction following capabilties even compared with larger LLMs The secret sauce? Cascade RL. 🧬 1️⃣ Cascade RL not only pushes the model limits on each domain, but also generates elite "teachers" for every expert domain. 2️⃣ Multi-domain on-policy distillation uses expert teachers to keep the student model sharp, mitigating domain shifts and matching expert-level performance. 💻 Read the blog: https://t.co/mqmSMXFsMx 📑 Check the paper: https://t.co/DRTloSE1h0 🤗 Get the weights: https://t.co/LvIo0WWgI5 #AI #MachineLearning #NVIDIA #LLM

Media 1
Media 2
View on Twitter

📊 Media Metadata

{
  "media": [
    {
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2036566157910745522/media_0.jpg",
      "media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2036566157910745522/media_0.jpg",
      "type": "photo",
      "filename": "media_0.jpg"
    },
    {
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2036566157910745522/media_2.jpg",
      "media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2036566157910745522/media_2.jpg",
      "type": "photo",
      "filename": "media_2.jpg"
    }
  ],
  "processed_at": "2026-03-25T01:50:37.137924",
  "pipeline_version": "2.0"
}

🔧 Raw API Response

{
  "type": "tweet",
  "id": "2036566157910745522",
  "url": "https://x.com/wbx_life/status/2036566157910745522",
  "twitterUrl": "https://twitter.com/wbx_life/status/2036566157910745522",
  "text": "🔥 The era of ultra-efficient reasoning is here: our Nemotron-Cascade-2-30B-A3B currently trending at #1 on huggingface 🤗\n\n🥇 Gold Medal-level performance on IMO 2025, IOI 2025, and ICPC World Finals 2025 -- all from a model with only 3B active parameters. 🤯\n\n⚡ SOTA alignment and instruction following capabilties even compared with larger LLMs\n\nThe secret sauce? Cascade RL. 🧬\n\n1️⃣ Cascade RL not only pushes the model limits on each domain, but also generates elite \"teachers\" for every expert domain. \n2️⃣ Multi-domain on-policy distillation uses expert teachers to keep the student model sharp, mitigating domain shifts and matching expert-level performance.\n\n💻 Read the blog: https://t.co/mqmSMXFsMx\n📑 Check the paper: https://t.co/DRTloSE1h0\n🤗 Get the weights: https://t.co/LvIo0WWgI5\n\n#AI #MachineLearning #NVIDIA #LLM",
  "source": "Twitter for iPhone",
  "retweetCount": 6,
  "replyCount": 2,
  "likeCount": 28,
  "quoteCount": 0,
  "viewCount": 1636,
  "createdAt": "Tue Mar 24 22:09:44 +0000 2026",
  "lang": "en",
  "bookmarkCount": 5,
  "isReply": false,
  "inReplyToId": null,
  "conversationId": "2036566157910745522",
  "displayTextRange": [
    0,
    274
  ],
  "inReplyToUserId": null,
  "inReplyToUsername": null,
  "author": {
    "type": "user",
    "userName": "wbx_life",
    "url": "https://x.com/wbx_life",
    "twitterUrl": "https://twitter.com/wbx_life",
    "id": "2311495605",
    "name": "Boxin Wang",
    "isVerified": false,
    "isBlueVerified": true,
    "verifiedType": null,
    "profilePicture": "https://pbs.twimg.com/profile_images/1912607007930216450/PJDVydUU_normal.jpg",
    "coverPicture": "https://pbs.twimg.com/profile_banners/2311495605/1589320315",
    "description": "",
    "location": "Bay Area, California, US",
    "followers": 843,
    "following": 528,
    "status": "",
    "canDm": false,
    "canMediaTag": true,
    "createdAt": "Wed Jan 29 08:49:18 +0000 2014",
    "entities": {
      "description": {
        "urls": []
      },
      "url": {}
    },
    "fastFollowersCount": 0,
    "favouritesCount": 512,
    "hasCustomTimelines": true,
    "isTranslator": false,
    "mediaCount": 4,
    "statusesCount": 110,
    "withheldInCountries": [],
    "affiliatesHighlightedLabel": {},
    "possiblySensitive": false,
    "pinnedTweetIds": [
      "2036566157910745522"
    ],
    "profile_bio": {
      "description": "Sr. Research Scientist @NVIDIA | UIUC Ph.D @IllinoisCS | LLM Post-training | Ex-intern at MSR @MSFTResearch, Google Research @googleai",
      "entities": {
        "description": {
          "hashtags": [],
          "symbols": [],
          "urls": [],
          "user_mentions": [
            {
              "id_str": "0",
              "indices": [
                23,
                30
              ],
              "name": "",
              "screen_name": "NVIDIA"
            },
            {
              "id_str": "0",
              "indices": [
                43,
                54
              ],
              "name": "",
              "screen_name": "IllinoisCS"
            },
            {
              "id_str": "0",
              "indices": [
                94,
                107
              ],
              "name": "",
              "screen_name": "MSFTResearch"
            },
            {
              "id_str": "0",
              "indices": [
                125,
                134
              ],
              "name": "",
              "screen_name": "googleai"
            }
          ]
        },
        "url": {
          "urls": [
            {
              "display_url": "boxin.wang",
              "expanded_url": "https://boxin.wang",
              "indices": [
                0,
                23
              ],
              "url": "https://t.co/3zJN9w5HXL"
            }
          ]
        }
      }
    },
    "isAutomated": false,
    "automatedBy": null
  },
  "extendedEntities": {
    "media": [
      {
        "display_url": "pic.twitter.com/B9wsH1cHoK",
        "expanded_url": "https://twitter.com/wbx_life/status/2036566157910745522/photo/1",
        "ext_media_availability": {
          "status": "Available"
        },
        "features": {
          "large": {
            "faces": []
          },
          "orig": {
            "faces": []
          }
        },
        "id_str": "2036563285085134848",
        "indices": [
          275,
          298
        ],
        "media_key": "3_2036563285085134848",
        "media_results": {
          "id": "QXBpTWVkaWFSZXN1bHRzOgwAAQoAARxDU4SCGiAACgACHENWIWPaIbIAAA==",
          "result": {
            "__typename": "ApiMedia",
            "id": "QXBpTWVkaWE6DAABCgABHENThIIaIAAKAAIcQ1YhY9ohsgAA",
            "media_key": "3_2036563285085134848"
          }
        },
        "media_url_https": "https://pbs.twimg.com/media/HENThIIaIAAvsyk.png",
        "original_info": {
          "focus_rects": [
            {
              "h": 317,
              "w": 566,
              "x": 0,
              "y": 0
            },
            {
              "h": 317,
              "w": 317,
              "x": 0,
              "y": 0
            },
            {
              "h": 317,
              "w": 278,
              "x": 0,
              "y": 0
            },
            {
              "h": 317,
              "w": 159,
              "x": 33,
              "y": 0
            },
            {
              "h": 317,
              "w": 641,
              "x": 0,
              "y": 0
            }
          ],
          "height": 317,
          "width": 641
        },
        "sizes": {
          "large": {
            "h": 317,
            "w": 641
          }
        },
        "type": "photo",
        "url": "https://t.co/B9wsH1cHoK"
      }
    ]
  },
  "card": null,
  "place": {},
  "entities": {
    "hashtags": [
      {
        "indices": [
          791,
          794
        ],
        "text": "AI"
      },
      {
        "indices": [
          795,
          811
        ],
        "text": "MachineLearning"
      },
      {
        "indices": [
          812,
          819
        ],
        "text": "NVIDIA"
      },
      {
        "indices": [
          820,
          824
        ],
        "text": "LLM"
      }
    ],
    "symbols": [],
    "urls": [
      {
        "display_url": "research.nvidia.com/labs/nemotron/…",
        "expanded_url": "https://research.nvidia.com/labs/nemotron/nemotron-cascade-2/",
        "indices": [
          680,
          703
        ],
        "url": "https://t.co/mqmSMXFsMx"
      },
      {
        "display_url": "arxiv.org/abs/2603.19220",
        "expanded_url": "https://arxiv.org/abs/2603.19220",
        "indices": [
          723,
          746
        ],
        "url": "https://t.co/DRTloSE1h0"
      },
      {
        "display_url": "huggingface.co/collections/nv…",
        "expanded_url": "https://huggingface.co/collections/nvidia/nemotron-cascade-2",
        "indices": [
          766,
          789
        ],
        "url": "https://t.co/LvIo0WWgI5"
      }
    ],
    "user_mentions": []
  },
  "quoted_tweet": {
    "type": "tweet",
    "id": "2036496268936224941",
    "url": "https://x.com/ctnzr/status/2036496268936224941",
    "twitterUrl": "https://twitter.com/ctnzr/status/2036496268936224941",
    "text": "Thank you to everyone in the community who is testing and using Nemotron models. It's great to see Nemotron-Cascade-2, Nemotron-3-Super and Nemotron-3-Nano trending on HF.\n\nThe Nemotron team is working hard to incorporate all your feedback into Nemotron 4.\n\nAnd yes, Nemotron 3 Ultra is still on track for release.\n\nhttps://t.co/lkEwmlUng9",
    "source": "Twitter for iPhone",
    "retweetCount": 28,
    "replyCount": 12,
    "likeCount": 140,
    "quoteCount": 7,
    "viewCount": 17540,
    "createdAt": "Tue Mar 24 17:32:01 +0000 2026",
    "lang": "en",
    "bookmarkCount": 18,
    "isReply": false,
    "inReplyToId": null,
    "conversationId": "2036496268936224941",
    "displayTextRange": [
      0,
      277
    ],
    "inReplyToUserId": null,
    "inReplyToUsername": null,
    "author": {
      "type": "user",
      "userName": "ctnzr",
      "url": "https://x.com/ctnzr",
      "twitterUrl": "https://twitter.com/ctnzr",
      "id": "257642411",
      "name": "Bryan Catanzaro",
      "isVerified": false,
      "isBlueVerified": true,
      "verifiedType": null,
      "profilePicture": "https://pbs.twimg.com/profile_images/1699513548647596032/QZ3dWSYh_normal.jpg",
      "coverPicture": "https://pbs.twimg.com/profile_banners/257642411/1710206605",
      "description": "",
      "location": "",
      "followers": 25367,
      "following": 474,
      "status": "",
      "canDm": true,
      "canMediaTag": false,
      "createdAt": "Fri Feb 25 21:39:13 +0000 2011",
      "entities": {
        "description": {
          "urls": []
        },
        "url": {}
      },
      "fastFollowersCount": 0,
      "favouritesCount": 4754,
      "hasCustomTimelines": true,
      "isTranslator": false,
      "mediaCount": 107,
      "statusesCount": 1713,
      "withheldInCountries": [],
      "affiliatesHighlightedLabel": {},
      "possiblySensitive": false,
      "pinnedTweetIds": [],
      "profile_bio": {
        "description": "VP, Applied Deep Learning Research @ NVIDIA",
        "entities": {
          "description": {
            "hashtags": [],
            "symbols": [],
            "urls": [],
            "user_mentions": []
          }
        }
      },
      "isAutomated": false,
      "automatedBy": null
    },
    "extendedEntities": {
      "media": [
        {
          "display_url": "pic.twitter.com/TBvyauSwCC",
          "expanded_url": "https://twitter.com/ctnzr/status/2036496268936224941/photo/1",
          "ext_media_availability": {
            "status": "Available"
          },
          "features": {
            "large": {
              "faces": []
            },
            "orig": {
              "faces": []
            }
          },
          "id_str": "2036495828110688256",
          "indices": [
            278,
            301
          ],
          "media_key": "3_2036495828110688256",
          "media_results": {
            "id": "QXBpTWVkaWFSZXN1bHRzOgwAAQoAARxDFip1GkAACgACHEMWkRhaIK0AAA==",
            "result": {
              "__typename": "ApiMedia",
              "id": "QXBpTWVkaWE6DAABCgABHEMWKnUaQAAKAAIcQxaRGFogrQAA",
              "media_key": "3_2036495828110688256"
            }
          },
          "media_url_https": "https://pbs.twimg.com/media/HEMWKnUaQAAwpwc.jpg",
          "original_info": {
            "focus_rects": [
              {
                "h": 1412,
                "w": 2521,
                "x": 561,
                "y": 0
              },
              {
                "h": 1412,
                "w": 1412,
                "x": 1522,
                "y": 0
              },
              {
                "h": 1412,
                "w": 1239,
                "x": 1609,
                "y": 0
              },
              {
                "h": 1412,
                "w": 706,
                "x": 1875,
                "y": 0
              },
              {
                "h": 1412,
                "w": 3082,
                "x": 0,
                "y": 0
              }
            ],
            "height": 1412,
            "width": 3082
          },
          "sizes": {
            "large": {
              "h": 938,
              "w": 2048
            }
          },
          "type": "photo",
          "url": "https://t.co/TBvyauSwCC"
        }
      ]
    },
    "card": null,
    "place": {},
    "entities": {
      "hashtags": [],
      "symbols": [],
      "urls": [
        {
          "display_url": "huggingface.co/models?pipelin…",
          "expanded_url": "https://huggingface.co/models?pipeline_tag=text-generation&sort=trending",
          "indices": [
            316,
            339
          ],
          "url": "https://t.co/lkEwmlUng9"
        }
      ],
      "user_mentions": []
    },
    "quoted_tweet": null,
    "retweeted_tweet": null,
    "isLimitedReply": false,
    "article": null
  },
  "retweeted_tweet": null,
  "isLimitedReply": false,
  "article": null
}