🐦 Twitter Post Details

Viewing enriched Twitter post

@omarsar0

// THE CASE FOR ENVIRONMENT SCALING // Environment scaling may be as important as model scaling for agentic AI. Current AI research suggests that building a powerful agentic AI model isn't just about better reasoning. It's also about better environments. The default approach to training capable AI agents today is collecting static trajectories or human demonstrations. This requires more data, more examples, and more annotation effort. But static data can't teach dynamic decision-making. Models trained this way struggle with the long-horizon, goal-oriented nature of real agentic tasks. This new research introduces Nex-N1, a framework that systematically scales the diversity and complexity of interactive training environments rather than just scaling data. Agent capabilities emerge from interaction, not imitation. Instead of collecting more demonstrations, they built infrastructure to automatically generate diverse agent architectures and workflows from natural language specifications. The system has three components. NexAU (Agent Universe) provides a universal agent framework that generates complex agent hierarchies from simple configurations. NexA4A (Agent for Agent) automatically synthesizes diverse agent architectures from natural language. NexGAP bridges the simulation-reality gap by integrating real-world MCP tools for grounded trajectory synthesis. Results: - On the τ2-bench, Nex-N1 built on DeepSeek-V3.1 scores 80.2, outperforming the base model's 42.8. - On SWE-bench Verified, Qwen3-32B-Nex-N1 achieves 50.5% compared to the base model's 12.9%. - On BFCL v4 for tool use, Nex-N1 (65.3) outperforms GPT-5 (61.6). In human evaluations on real-world project development across 43 coding scenarios, Nex-N1 wins or ties against Claude Sonnet 4.5 in 64.5% of cases and against GPT-5 in ~70% of cases. They also built a deep research agent on Nex-N1, achieving 47.0% on the Deep Research Benchmark, with capabilities for visualized report generation, including slides and research posters. Paper: https://t.co/Ny7G15XEwi

Media 1
View on Twitter

📊 Media Metadata

{
  "media": [
    {
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/1997329447691964626/media_0.jpg?",
      "media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/1997329447691964626/media_0.jpg?",
      "type": "photo",
      "filename": "media_0.jpg"
    }
  ],
  "processed_at": "2025-12-08T13:23:13.479107",
  "pipeline_version": "2.0"
}

🔧 Raw API Response

{
  "type": "tweet",
  "id": "1997329447691964626",
  "url": "https://x.com/omarsar0/status/1997329447691964626",
  "twitterUrl": "https://twitter.com/omarsar0/status/1997329447691964626",
  "text": "// THE CASE FOR ENVIRONMENT SCALING //\n\nEnvironment scaling may be as important as model scaling for agentic AI.\n\nCurrent AI research suggests that building a powerful agentic AI model isn't just about better reasoning. It's also about better environments.\n\nThe default approach to training capable AI agents today is collecting static trajectories or human demonstrations. This requires more data, more examples, and more annotation effort.\n\nBut static data can't teach dynamic decision-making. Models trained this way struggle with the long-horizon, goal-oriented nature of real agentic tasks.\n\nThis new research introduces Nex-N1, a framework that systematically scales the diversity and complexity of interactive training environments rather than just scaling data.\n\nAgent capabilities emerge from interaction, not imitation. Instead of collecting more demonstrations, they built infrastructure to automatically generate diverse agent architectures and workflows from natural language specifications.\n\nThe system has three components. NexAU (Agent Universe) provides a universal agent framework that generates complex agent hierarchies from simple configurations. NexA4A (Agent for Agent) automatically synthesizes diverse agent architectures from natural language. NexGAP bridges the simulation-reality gap by integrating real-world MCP tools for grounded trajectory synthesis.\n\nResults:\n\n- On the τ2-bench, Nex-N1 built on DeepSeek-V3.1 scores 80.2, outperforming the base model's 42.8.\n- On SWE-bench Verified, Qwen3-32B-Nex-N1 achieves 50.5% compared to the base model's 12.9%.\n- On BFCL v4 for tool use, Nex-N1 (65.3) outperforms GPT-5 (61.6).\n\nIn human evaluations on real-world project development across 43 coding scenarios, Nex-N1 wins or ties against Claude Sonnet 4.5 in 64.5% of cases and against GPT-5 in ~70% of cases.\n\nThey also built a deep research agent on Nex-N1, achieving 47.0% on the Deep Research Benchmark, with capabilities for visualized report generation, including slides and research posters.\n\nPaper: https://t.co/Ny7G15XEwi",
  "source": "Twitter for iPhone",
  "retweetCount": 13,
  "replyCount": 8,
  "likeCount": 99,
  "quoteCount": 1,
  "viewCount": 10157,
  "createdAt": "Sat Dec 06 15:37:03 +0000 2025",
  "lang": "en",
  "bookmarkCount": 61,
  "isReply": false,
  "inReplyToId": null,
  "conversationId": "1997329447691964626",
  "displayTextRange": [
    0,
    279
  ],
  "inReplyToUserId": null,
  "inReplyToUsername": null,
  "author": {
    "type": "user",
    "userName": "omarsar0",
    "url": "https://x.com/omarsar0",
    "twitterUrl": "https://twitter.com/omarsar0",
    "id": "3448284313",
    "name": "elvis",
    "isVerified": false,
    "isBlueVerified": true,
    "verifiedType": null,
    "profilePicture": "https://pbs.twimg.com/profile_images/939313677647282181/vZjFWtAn_normal.jpg",
    "coverPicture": "https://pbs.twimg.com/profile_banners/3448284313/1565974901",
    "description": "Building @dair_ai • Ex Meta AI, Elastic, PhD • New cohort: https://t.co/xw2XQ0z8up",
    "location": "DAIR.AI Academy",
    "followers": 278411,
    "following": 727,
    "status": "",
    "canDm": true,
    "canMediaTag": false,
    "createdAt": "Fri Sep 04 12:59:26 +0000 2015",
    "entities": {
      "description": {
        "urls": [
          {
            "display_url": "dair-ai.thinkific.com/courses/buildi…",
            "expanded_url": "https://dair-ai.thinkific.com/courses/building-effective-ai-agents-2",
            "url": "https://t.co/xw2XQ0z8up",
            "indices": [
              59,
              82
            ]
          }
        ]
      },
      "url": {
        "urls": [
          {
            "display_url": "dair.ai",
            "expanded_url": "https://www.dair.ai/",
            "url": "https://t.co/XQto5ypkSM",
            "indices": [
              0,
              23
            ]
          }
        ]
      }
    },
    "fastFollowersCount": 0,
    "favouritesCount": 33796,
    "hasCustomTimelines": true,
    "isTranslator": false,
    "mediaCount": 4362,
    "statusesCount": 16685,
    "withheldInCountries": [],
    "affiliatesHighlightedLabel": {},
    "possiblySensitive": false,
    "pinnedTweetIds": [
      "1997717251546583103"
    ],
    "profile_bio": {},
    "isAutomated": false,
    "automatedBy": null
  },
  "extendedEntities": {
    "media": [
      {
        "display_url": "pic.x.com/exkBMa95sc",
        "expanded_url": "https://x.com/omarsar0/status/1997329447691964626/photo/1",
        "id_str": "1997329444240027648",
        "indices": [
          280,
          303
        ],
        "media_key": "3_1997329444240027648",
        "media_url_https": "https://pbs.twimg.com/media/G7fwjBda8AAAV-p.jpg",
        "type": "photo",
        "url": "https://t.co/exkBMa95sc",
        "ext_media_availability": {
          "status": "Available"
        },
        "features": {
          "large": {
            "faces": []
          },
          "medium": {
            "faces": []
          },
          "small": {
            "faces": []
          },
          "orig": {
            "faces": []
          }
        },
        "sizes": {
          "large": {
            "h": 1758,
            "w": 1618,
            "resize": "fit"
          },
          "medium": {
            "h": 1200,
            "w": 1104,
            "resize": "fit"
          },
          "small": {
            "h": 680,
            "w": 626,
            "resize": "fit"
          },
          "thumb": {
            "h": 150,
            "w": 150,
            "resize": "crop"
          }
        },
        "original_info": {
          "height": 1758,
          "width": 1618,
          "focus_rects": [
            {
              "x": 0,
              "y": 0,
              "w": 1618,
              "h": 906
            },
            {
              "x": 0,
              "y": 0,
              "w": 1618,
              "h": 1618
            },
            {
              "x": 0,
              "y": 0,
              "w": 1542,
              "h": 1758
            },
            {
              "x": 0,
              "y": 0,
              "w": 879,
              "h": 1758
            },
            {
              "x": 0,
              "y": 0,
              "w": 1618,
              "h": 1758
            }
          ]
        },
        "media_results": {
          "result": {
            "media_key": "3_1997329444240027648"
          }
        }
      }
    ]
  },
  "card": null,
  "place": {},
  "entities": {
    "hashtags": [],
    "symbols": [],
    "urls": [
      {
        "display_url": "arxiv.org/abs/2512.04987",
        "expanded_url": "https://arxiv.org/abs/2512.04987",
        "url": "https://t.co/Ny7G15XEwi",
        "indices": [
          2034,
          2057
        ]
      }
    ],
    "user_mentions": []
  },
  "quoted_tweet": null,
  "retweeted_tweet": null,
  "article": null
}