🐦 Twitter Post Details

Viewing enriched Twitter post

@s_batzoglou

I evaluated Ox Alpha on the induction benchmark, accessing the model through OpenRouter. The performance of Ox Alpha on this benchmark is OK but not great. It ranks below Luna and DeepSeek v4 Pro and just above Gemini 3.7 Flash. I have seen other benchmarks posted on X where Ox Alpha rocks, so this could be very benchmark-specific. It was not easy to get evaluable results (at least on OpenRouter). Ox Alpha often returned "" in responses, or API errors. I made 551 API calls to be able to get a reasonable completion rate of 87/100. I wonder if this is an OpenRouter issue or something broader like the model failing to return answers when it hasn't solved the problem. This is very common with reasoning models, but more pronounced than average here. Some words on the induction benchmark: This is a challenging reasoning benchmark, described in ICML 2026. The models are given several small graphs in which some nodes are marked as targets. The task is to provide a first-order logical formula that picks precisely the target nodes in all graphs simultaneously. Correct: a formula that correctly picks precisely the marked nodes. Holdout correct: a formula that correctly picks precisely the marked nodes in held out problems. Formula complexity (in AST): the avg tree size of the correct formula. GitHub public repository: https://t.co/ZXXqzouuni Paper: https://t.co/gBelIZQEaa

Media 1

📊 Media Metadata

{
  "media": [
    {
      "type": "photo",
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2091147248197406978/media_0.jpg",
      "filename": "media_0.jpg"
    }
  ],
  "processed_at": "2026-08-22T13:01:34.068755",
  "pipeline_version": "2.0"
}

🔧 Raw API Response

{
  "type": "tweet",
  "id": "2091147248197406978",
  "url": "https://x.com/s_batzoglou/status/2091147248197406978",
  "twitterUrl": "https://twitter.com/s_batzoglou/status/2091147248197406978",
  "text": "I evaluated Ox Alpha on the induction benchmark, accessing the model through OpenRouter.\n\nThe performance of Ox Alpha on this benchmark is OK but not great. It ranks below Luna and DeepSeek v4 Pro and just above Gemini 3.7 Flash. I have seen other benchmarks posted on X where Ox Alpha rocks, so this could be very benchmark-specific.\n\nIt was not easy to get evaluable results (at least on OpenRouter). Ox Alpha often returned \"\" in responses, or API errors. I made 551 API calls to be able to get a reasonable completion rate of 87/100. I wonder if this is an OpenRouter issue or something broader like the model failing to return answers when it hasn't solved the problem. This is very common with reasoning models, but more pronounced than average here.\n\nSome words on the induction benchmark:\n\nThis is a challenging reasoning benchmark, described in ICML 2026.  The models are given several small graphs in which some nodes are marked as targets. \n\nThe task is to provide a first-order logical formula that picks precisely the target nodes in all graphs simultaneously.  \n\nCorrect: a formula that correctly picks precisely the marked nodes. \n\nHoldout correct: a formula that correctly picks precisely the marked nodes in held out problems.  \n\nFormula complexity (in AST): the avg tree size of the correct formula.  \n\nGitHub public repository: https://t.co/ZXXqzouuni\n\nPaper: https://t.co/gBelIZQEaa",
  "source": "Twitter for iPhone",
  "retweetCount": 0,
  "replyCount": 1,
  "likeCount": 1,
  "quoteCount": 0,
  "viewCount": 51,
  "createdAt": "Sat Aug 22 12:55:30 +0000 2026",
  "lang": "en",
  "bookmarkCount": 0,
  "isReply": false,
  "inReplyToId": null,
  "conversationId": "2091147248197406978",
  "displayTextRange": [
    0,
    279
  ],
  "inReplyToUserId": null,
  "inReplyToUsername": null,
  "author": {
    "type": "user",
    "userName": "s_batzoglou",
    "url": "https://x.com/s_batzoglou",
    "twitterUrl": "https://twitter.com/s_batzoglou",
    "id": "1518735949458378752",
    "name": "Serafim Batzoglou",
    "isVerified": false,
    "isBlueVerified": true,
    "verifiedType": null,
    "profilePicture": "https://pbs.twimg.com/profile_images/1518736918527152128/hV7H_k58_normal.jpg",
    "coverPicture": "https://pbs.twimg.com/profile_banners/1518735949458378752/1786275261",
    "description": "",
    "location": "San Francisco and Miami",
    "followers": 3466,
    "following": 884,
    "status": "",
    "canDm": false,
    "canMediaTag": true,
    "createdAt": "Mon Apr 25 23:37:36 +0000 2022",
    "entities": {
      "description": {
        "urls": []
      },
      "url": {}
    },
    "fastFollowersCount": 0,
    "favouritesCount": 50429,
    "hasCustomTimelines": true,
    "isTranslator": false,
    "mediaCount": 332,
    "statusesCount": 6808,
    "withheldInCountries": [],
    "affiliatesHighlightedLabel": {},
    "possiblySensitive": false,
    "pinnedTweetIds": [],
    "profile_bio": {
      "description": "Welcome to the Singularity. AI - biotech - math - philosophy of mind. CSO @seer_bio; former prof @StanfordAILab; cofounder @dnanexus; opinions entirely my own",
      "entities": {
        "description": {
          "user_mentions": [
            {
              "id_str": "",
              "indices": [
                74,
                83
              ],
              "name": "",
              "screen_name": "seer_bio"
            },
            {
              "id_str": "",
              "indices": [
                97,
                111
              ],
              "name": "",
              "screen_name": "StanfordAILab"
            },
            {
              "id_str": "",
              "indices": [
                123,
                132
              ],
              "name": "",
              "screen_name": "dnanexus"
            }
          ]
        }
      }
    },
    "isAutomated": false,
    "automatedBy": null
  },
  "extendedEntities": {
    "media": [
      {
        "allow_download_status": {
          "allow_download": true
        },
        "display_url": "pic.x.com/WQTaMOe2Ii",
        "expanded_url": "https://x.com/s_batzoglou/status/2091147248197406978/photo/1",
        "ext_master_playlist_only": [],
        "ext_media_availability": {
          "status": "Available"
        },
        "ext_playlists": [],
        "features": {
          "large": {
            "faces": []
          },
          "orig": {
            "faces": []
          }
        },
        "id_str": "2091145311150358528",
        "indices": [
          280,
          303
        ],
        "media_key": "3_2091145311150358528",
        "media_results": {
          "id": "QXBpTWVkaWFSZXN1bHRzOgwAAQoAAR0FPZTE10AACgACHQU/V8XXkQIAAA==",
          "result": {
            "__typename": "ApiMedia",
            "id": "QXBpTWVkaWE6DAABCgABHQU9lMTXQAAKAAIdBT9XxdeRAgAA",
            "media_key": "3_2091145311150358528"
          }
        },
        "media_url_https": "https://pbs.twimg.com/media/HQU9lMTXQAAPLZC.jpg",
        "original_info": {
          "focus_rects": [
            {
              "h": 937,
              "w": 1674,
              "x": 0,
              "y": 76
            },
            {
              "h": 1534,
              "w": 1534,
              "x": 0,
              "y": 0
            },
            {
              "h": 1534,
              "w": 1346,
              "x": 38,
              "y": 0
            },
            {
              "h": 1534,
              "w": 767,
              "x": 328,
              "y": 0
            },
            {
              "h": 1534,
              "w": 1674,
              "x": 0,
              "y": 0
            }
          ],
          "height": 1534,
          "width": 1674
        },
        "sizes": {
          "large": {
            "h": 1534,
            "w": 1674
          }
        },
        "type": "photo",
        "url": "https://t.co/WQTaMOe2Ii"
      }
    ]
  },
  "card": null,
  "place": {},
  "entities": {
    "hashtags": [],
    "symbols": [],
    "urls": [
      {
        "display_url": "github.com/SerafimBatzogl…",
        "expanded_url": "https://github.com/SerafimBatzoglou/concept-synth",
        "indices": [
          1347,
          1370
        ],
        "url": "https://t.co/ZXXqzouuni"
      },
      {
        "display_url": "arxiv.org/abs/2602.18956",
        "expanded_url": "https://arxiv.org/abs/2602.18956",
        "indices": [
          1379,
          1402
        ],
        "url": "https://t.co/gBelIZQEaa"
      }
    ],
    "user_mentions": []
  },
  "quoted_tweet": null,
  "retweeted_tweet": null,
  "isLimitedReply": false,
  "communityInfo": null,
  "article": null
}