🐦 Twitter Post Details

Viewing enriched Twitter post

@atomic_chat_hq

Diffusion Gemma is 4x faster, but makes 6x more mistakes! We benchmarked the new diffusion LLM against its autoregressive twin on a single H100 (FP8). We gave each the same three tasks: write a Steve Jobs biography, the history of Tetris, and the story of BeOS - every next topic less popular than the previous one. Then we fact-checked every claim in every answer. Gemma4 got 45 facts right, 5 wrong. DiffusionGemma got 33 right, 28 wrong. The less popular the topic, the worse it got: 4 mistakes on Jobs, 12 on Tetris, 12 on BeOS. It named Clara Clley as Steve Jobs' mother, invented a colleague for Pajitnov named Geri Gulovik and priced the BeBox at $9,999. The real one cost $1,600. Outputs: Gemma4 26B A4B: 218 tok/s · 15.1s total · 45 facts · 5 mistakes DiffusionGemma 26B A4B: 763 tok/s · 3.7s total · 33 facts · 28 mistakes The reason is simple. DiffusionGemma throws 256 tokens on the screen at once and polishes them pass after pass until the text sounds smooth. Smooth is all it cares about: a fake name, date or number sounds just as smooth as a real one, so it stays. Regular Gemma4 meanwhile writes one word at a time and checks every new word against everything before it. Google says it themselves in the launch post: quality is lower, use regular Gemma 4 when facts matter.

📊 Media Metadata

{
  "media": [
    {
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2065228400764063899/media_0.mp4",
      "media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2065228400764063899/media_0.mp4",
      "type": "video",
      "filename": "media_0.mp4"
    }
  ],
  "processed_at": "2026-06-12T08:46:10.172677",
  "pipeline_version": "2.0"
}

🔧 Raw API Response

{
  "type": "tweet",
  "id": "2065228400764063899",
  "url": "https://x.com/atomic_chat_hq/status/2065228400764063899",
  "twitterUrl": "https://twitter.com/atomic_chat_hq/status/2065228400764063899",
  "text": "Diffusion Gemma is 4x faster, but makes 6x more mistakes!\n\nWe benchmarked the new diffusion LLM against its autoregressive twin on a single H100 (FP8). We gave each the same three tasks: write a Steve Jobs biography, the history of Tetris, and the story of BeOS - every next topic less popular than the previous one. Then we fact-checked every claim in every answer.\n\nGemma4 got 45 facts right, 5 wrong. DiffusionGemma got 33 right, 28 wrong. The less popular the topic, the worse it got: 4 mistakes on Jobs, 12 on Tetris, 12 on BeOS. It named Clara Clley as Steve Jobs' mother, invented a colleague for Pajitnov named Geri Gulovik and priced the BeBox at $9,999. The real one cost $1,600.\n\nOutputs:\nGemma4 26B A4B: 218 tok/s · 15.1s total · 45 facts · 5 mistakes\nDiffusionGemma 26B A4B: 763 tok/s · 3.7s total · 33 facts · 28 mistakes\n\nThe reason is simple. DiffusionGemma throws 256 tokens on the screen at once and polishes them pass after pass until the text sounds smooth. Smooth is all it cares about: a fake name, date or number sounds just as smooth as a real one, so it stays. Regular Gemma4 meanwhile writes one word at a time and checks every new word against everything before it. Google says it themselves in the launch post: quality is lower, use regular Gemma 4 when facts matter.",
  "source": "Twitter for iPhone",
  "retweetCount": 26,
  "replyCount": 18,
  "likeCount": 242,
  "quoteCount": 14,
  "viewCount": 23217,
  "createdAt": "Fri Jun 12 00:23:15 +0000 2026",
  "lang": "en",
  "bookmarkCount": 90,
  "isReply": false,
  "inReplyToId": null,
  "conversationId": "2065228400764063899",
  "displayTextRange": [
    0,
    280
  ],
  "inReplyToUserId": null,
  "inReplyToUsername": null,
  "author": {
    "type": "user",
    "userName": "atomic_chat_hq",
    "url": "https://x.com/atomic_chat_hq",
    "twitterUrl": "https://twitter.com/atomic_chat_hq",
    "id": "2037196634489077761",
    "name": "atomic.chat",
    "isVerified": false,
    "isBlueVerified": true,
    "verifiedType": null,
    "profilePicture": "https://pbs.twimg.com/profile_images/2037454994899959808/L7uqgUfL_normal.jpg",
    "coverPicture": "https://pbs.twimg.com/profile_banners/2037196634489077761/1774602079",
    "description": "",
    "location": "California, USA",
    "followers": 8048,
    "following": 24,
    "status": "",
    "canDm": false,
    "canMediaTag": true,
    "createdAt": "Thu Mar 26 15:55:07 +0000 2026",
    "entities": {
      "description": {
        "urls": []
      },
      "url": {}
    },
    "fastFollowersCount": 0,
    "favouritesCount": 500,
    "hasCustomTimelines": true,
    "isTranslator": false,
    "mediaCount": 25,
    "statusesCount": 153,
    "withheldInCountries": [],
    "affiliatesHighlightedLabel": {},
    "possiblySensitive": false,
    "pinnedTweetIds": [
      "2062670255818956993"
    ],
    "profile_bio": {
      "description": "Local AI chat and Inference Engine. Enhanced by TurboQuant. Team: @gladkos @skinbagwbones @AlexFromAtomic @danyurkin",
      "entities": {
        "description": {
          "user_mentions": [
            {
              "id_str": "",
              "indices": [
                66,
                74
              ],
              "name": "",
              "screen_name": "gladkos"
            },
            {
              "id_str": "",
              "indices": [
                75,
                89
              ],
              "name": "",
              "screen_name": "skinbagwbones"
            },
            {
              "id_str": "",
              "indices": [
                90,
                105
              ],
              "name": "",
              "screen_name": "AlexFromAtomic"
            },
            {
              "id_str": "",
              "indices": [
                106,
                116
              ],
              "name": "",
              "screen_name": "danyurkin"
            }
          ]
        },
        "url": {
          "urls": [
            {
              "display_url": "atomic.chat",
              "expanded_url": "http://atomic.chat",
              "indices": [
                0,
                23
              ],
              "url": "https://t.co/lW9A0SM1LR"
            }
          ]
        }
      }
    },
    "isAutomated": false,
    "automatedBy": null
  },
  "extendedEntities": {
    "media": [
      {
        "additional_media_info": {
          "monetizable": true
        },
        "allow_download_status": {
          "allow_download": true
        },
        "display_url": "pic.twitter.com/U1GsZ6Z8wI",
        "expanded_url": "https://twitter.com/atomic_chat_hq/status/2065228400764063899/video/1",
        "ext_master_playlist_only": [],
        "ext_media_availability": {
          "status": "Available"
        },
        "ext_playlists": [],
        "id_str": "2065228296032329728",
        "indices": [
          281,
          304
        ],
        "media_key": "13_2065228296032329728",
        "media_results": {
          "id": "QXBpTWVkaWFSZXN1bHRzOgwABAoAARypKjF1F6AAAAA=",
          "result": {
            "__typename": "ApiMedia",
            "id": "QXBpTWVkaWE6DAAECgABHKkqMXUXoAAAAA==",
            "media_key": "13_2065228296032329728"
          }
        },
        "media_url_https": "https://pbs.twimg.com/amplify_video_thumb/2065228296032329728/img/cFIc7IevA286PFnt.jpg",
        "original_info": {
          "focus_rects": [],
          "height": 1080,
          "width": 1888
        },
        "sizes": {
          "large": {
            "h": 1080,
            "w": 1888
          }
        },
        "type": "video",
        "url": "https://t.co/U1GsZ6Z8wI",
        "video_info": {
          "aspect_ratio": [
            236,
            135
          ],
          "duration_millis": 32216,
          "variants": [
            {
              "content_type": "application/x-mpegURL",
              "url": "https://video.twimg.com/amplify_video/2065228296032329728/pl/O1xlvBaVA958IUfn.m3u8?tag=27"
            },
            {
              "bitrate": 256000,
              "content_type": "video/mp4",
              "url": "https://video.twimg.com/amplify_video/2065228296032329728/vid/avc1/472x270/KnIDNtpcA3DdRvB5.mp4?tag=27"
            },
            {
              "bitrate": 832000,
              "content_type": "video/mp4",
              "url": "https://video.twimg.com/amplify_video/2065228296032329728/vid/avc1/628x360/RjYf8WRkNyXvqoUF.mp4?tag=27"
            },
            {
              "bitrate": 2176000,
              "content_type": "video/mp4",
              "url": "https://video.twimg.com/amplify_video/2065228296032329728/vid/avc1/1258x720/2kpKoJDr6-8pzmtO.mp4?tag=27"
            },
            {
              "bitrate": 10368000,
              "content_type": "video/mp4",
              "url": "https://video.twimg.com/amplify_video/2065228296032329728/vid/avc1/1888x1080/_teEAmpXYA5e5CHP.mp4?tag=27"
            }
          ]
        }
      }
    ]
  },
  "card": null,
  "place": {},
  "entities": {
    "hashtags": [],
    "symbols": [],
    "timestamps": [],
    "urls": [],
    "user_mentions": []
  },
  "quoted_tweet": null,
  "retweeted_tweet": null,
  "isLimitedReply": false,
  "communityInfo": null,
  "article": null
}