Viewing enriched Twitter post
This guy is BEYOND CRACKED. Gemma 4 already on MLX, bro has uploaded all models with quantization. 125 models uploaded in last few hours š¤Æ New mlx-vlm repo also supports turbo-quant, and rf-detr too (among other things) If you are a mac dev, you better be jumping at this. Bookmark him, turn his notifications on, sponsor his work.
{
"media": [
{
"url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2039821196557164656/media_0.jpg",
"media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2039821196557164656/media_0.jpg",
"type": "photo",
"filename": "media_0.jpg"
}
],
"processed_at": "2026-04-04T00:20:34.893780",
"pipeline_version": "2.0"
}{
"type": "tweet",
"id": "2039821196557164656",
"url": "https://x.com/neural_avb/status/2039821196557164656",
"twitterUrl": "https://twitter.com/neural_avb/status/2039821196557164656",
"text": "This guy is BEYOND CRACKED.\n\nGemma 4 already on MLX, bro has uploaded all models with quantization. 125 models uploaded in last few hours š¤Æ\n\nNew mlx-vlm repo also supports turbo-quant, and rf-detr too (among other things)\n\nIf you are a mac dev, you better be jumping at this. Bookmark him, turn his notifications on, sponsor his work.",
"source": "Twitter for iPhone",
"retweetCount": 230,
"replyCount": 59,
"likeCount": 3591,
"quoteCount": 7,
"viewCount": 479222,
"createdAt": "Thu Apr 02 21:44:05 +0000 2026",
"lang": "en",
"bookmarkCount": 4714,
"isReply": false,
"inReplyToId": null,
"conversationId": "2039821196557164656",
"displayTextRange": [
0,
275
],
"inReplyToUserId": null,
"inReplyToUsername": null,
"author": {
"type": "user",
"userName": "neural_avb",
"url": "https://x.com/neural_avb",
"twitterUrl": "https://twitter.com/neural_avb",
"id": "1754194661084983296",
"name": "AVB",
"isVerified": false,
"isBlueVerified": true,
"verifiedType": null,
"profilePicture": "https://pbs.twimg.com/profile_images/2015375309147611136/WKvfQ-oV_normal.jpg",
"coverPicture": "https://pbs.twimg.com/profile_banners/1754194661084983296/1769442083",
"description": "",
"location": "",
"followers": 8872,
"following": 349,
"status": "",
"canDm": true,
"canMediaTag": true,
"createdAt": "Sun Feb 04 17:26:41 +0000 2024",
"entities": {
"description": {
"urls": []
},
"url": {}
},
"fastFollowersCount": 0,
"favouritesCount": 5370,
"hasCustomTimelines": true,
"isTranslator": false,
"mediaCount": 1373,
"statusesCount": 5218,
"withheldInCountries": [],
"affiliatesHighlightedLabel": {},
"possiblySensitive": false,
"pinnedTweetIds": [
"2038371829266743788"
],
"profile_bio": {
"description": "Not 21 | Neural Breakdown on YT | Read research papers with AI: https://t.co/T4MlMfP1oP | Latest vid: RLMs, CPT | Next: SFT, Synthetic data gen",
"entities": {
"description": {
"hashtags": [],
"symbols": [],
"urls": [
{
"display_url": "paperbreakdown.com",
"expanded_url": "https://paperbreakdown.com",
"indices": [
64,
87
],
"url": "https://t.co/T4MlMfP1oP"
}
],
"user_mentions": []
},
"url": {
"urls": [
{
"display_url": "youtube.com/@avb_fj",
"expanded_url": "http://youtube.com/@avb_fj",
"indices": [
0,
23
],
"url": "https://t.co/dSiHoIiCPh"
}
]
}
}
},
"isAutomated": false,
"automatedBy": null
},
"extendedEntities": {
"media": [
{
"allow_download_status": {
"allow_download": true
},
"display_url": "pic.twitter.com/PirPCvEeDa",
"expanded_url": "https://twitter.com/neural_avb/status/2039821196557164656/photo/1",
"ext_media_availability": {
"status": "Available"
},
"features": {
"large": {
"faces": [
{
"h": 62,
"w": 62,
"x": 468,
"y": 20
}
]
},
"orig": {
"faces": [
{
"h": 62,
"w": 62,
"x": 468,
"y": 20
}
]
}
},
"id_str": "2039819389957197824",
"indices": [
276,
299
],
"media_key": "3_2039819389957197824",
"media_results": {
"id": "QXBpTWVkaWFSZXN1bHRzOgwAAQoAARxO5O2AWlAACgACHE7mkiIaIHAAAA==",
"result": {
"__typename": "ApiMedia",
"id": "QXBpTWVkaWE6DAABCgABHE7k7YBaUAAKAAIcTuaSIhogcAAA",
"media_key": "3_2039819389957197824"
}
},
"media_url_https": "https://pbs.twimg.com/media/HE7k7YBaUAACfNz.jpg",
"original_info": {
"focus_rects": [
{
"h": 601,
"w": 1074,
"x": 0,
"y": 0
},
{
"h": 796,
"w": 796,
"x": 0,
"y": 0
},
{
"h": 796,
"w": 698,
"x": 0,
"y": 0
},
{
"h": 796,
"w": 398,
"x": 42,
"y": 0
},
{
"h": 796,
"w": 1074,
"x": 0,
"y": 0
}
],
"height": 796,
"width": 1074
},
"sizes": {
"large": {
"h": 796,
"w": 1074
}
},
"type": "photo",
"url": "https://t.co/PirPCvEeDa"
}
]
},
"card": null,
"place": {},
"entities": {
"hashtags": [],
"symbols": [],
"urls": [],
"user_mentions": []
},
"quoted_tweet": {
"type": "tweet",
"id": "2039815307821199709",
"url": "https://x.com/Prince_Canuma/status/2039815307821199709",
"twitterUrl": "https://twitter.com/Prince_Canuma/status/2039815307821199709",
"text": "mlx-vlm v0.4.3 is here š\n\nDay-0 support:\n š„ Gemma 4 (vision, audio, MoE) by @GoogleDeepMind \nš¦
Falcon-OCR + Falcon Perception by @TIIuae \nšŖØ Granite Vision 4.0 by @IBMResearch \n\nNew models: \nšÆ SAM 3.1 with Object Multiplex by @facebook \nš RF-DETR detection & segmentation by @roboflow \n\nInfra: \nā” TurboQuant (KV cache compression) \nš„ļø CUDA support for vision models (Sam and RF-DETR)\n\nGet started today:\n\n> uv pip install -U mlx-vlm\n\nLeave us a star āļø\nhttps://t.co/7BvnEuzKvj",
"source": "Twitter for iPhone",
"retweetCount": 166,
"replyCount": 67,
"likeCount": 1711,
"quoteCount": 26,
"viewCount": 664920,
"createdAt": "Thu Apr 02 21:20:41 +0000 2026",
"lang": "en",
"bookmarkCount": 2847,
"isReply": false,
"inReplyToId": null,
"conversationId": "2039815307821199709",
"displayTextRange": [
0,
277
],
"inReplyToUserId": null,
"inReplyToUsername": null,
"author": {
"type": "user",
"userName": "Prince_Canuma",
"url": "https://x.com/Prince_Canuma",
"twitterUrl": "https://twitter.com/Prince_Canuma",
"id": "726009698",
"name": "Prince Canuma",
"isVerified": false,
"isBlueVerified": true,
"verifiedType": null,
"profilePicture": "https://pbs.twimg.com/profile_images/2018690435636510720/Aq3-a3Mv_normal.jpg",
"coverPicture": "https://pbs.twimg.com/profile_banners/726009698/1770128291",
"description": "",
"location": "Krakow, Poland",
"followers": 18433,
"following": 1199,
"status": "",
"canDm": true,
"canMediaTag": true,
"createdAt": "Mon Jul 30 12:54:47 +0000 2012",
"entities": {
"description": {
"urls": []
},
"url": {}
},
"fastFollowersCount": 0,
"favouritesCount": 21186,
"hasCustomTimelines": true,
"isTranslator": false,
"mediaCount": 2612,
"statusesCount": 21438,
"withheldInCountries": [],
"affiliatesHighlightedLabel": {},
"possiblySensitive": false,
"pinnedTweetIds": [
"2026022469010599954"
],
"profile_bio": {
"description": "Apple MLX King š¤“š½ā¢ Creator of (mlx-audio & mlx-vlm) ā¢ Ex-@arcee_ai ā¢ @neptune_ai ā¢ https://t.co/iZnxoefJBU",
"entities": {
"description": {
"hashtags": [],
"symbols": [],
"urls": [
{
"display_url": "linktr.ee/prince.canuma",
"expanded_url": "https://linktr.ee/prince.canuma",
"indices": [
83,
106
],
"url": "https://t.co/iZnxoefJBU"
}
],
"user_mentions": [
{
"id_str": "0",
"indices": [
57,
66
],
"name": "",
"screen_name": "arcee_ai"
},
{
"id_str": "0",
"indices": [
69,
80
],
"name": "",
"screen_name": "neptune_ai"
}
]
},
"url": {
"urls": [
{
"display_url": "github.com/Blaizzy",
"expanded_url": "https://github.com/Blaizzy",
"indices": [
0,
23
],
"url": "https://t.co/ELxUGGEMjU"
}
]
}
}
},
"isAutomated": false,
"automatedBy": null
},
"extendedEntities": {
"media": [
{
"allow_download_status": {
"allow_download": true
},
"display_url": "pic.twitter.com/R6aWzXD0yg",
"expanded_url": "https://twitter.com/Prince_Canuma/status/2039815307821199709/photo/1",
"ext_media_availability": {
"status": "Available"
},
"features": {
"large": {
"faces": [
{
"h": 152,
"w": 152,
"x": 66,
"y": 367
},
{
"h": 213,
"w": 213,
"x": 401,
"y": 635
}
]
},
"orig": {
"faces": [
{
"h": 152,
"w": 152,
"x": 66,
"y": 367
},
{
"h": 213,
"w": 213,
"x": 401,
"y": 635
}
]
}
},
"id_str": "2039815114870939648",
"indices": [
278,
301
],
"media_key": "3_2039815114870939648",
"media_results": {
"id": "QXBpTWVkaWFSZXN1bHRzOgwAAQoAARxO4QohWuAACgACHE7hNw4WMV0AAA==",
"result": {
"__typename": "ApiMedia",
"id": "QXBpTWVkaWE6DAABCgABHE7hCiFa4AAKAAIcTuE3DhYxXQAA",
"media_key": "3_2039815114870939648"
}
},
"media_url_https": "https://pbs.twimg.com/media/HE7hCiFa4AASoya.jpg",
"original_info": {
"focus_rects": [
{
"h": 893,
"w": 1594,
"x": 0,
"y": 0
},
{
"h": 1548,
"w": 1548,
"x": 0,
"y": 0
},
{
"h": 1548,
"w": 1358,
"x": 0,
"y": 0
},
{
"h": 1548,
"w": 774,
"x": 0,
"y": 0
},
{
"h": 1548,
"w": 1594,
"x": 0,
"y": 0
}
],
"height": 1548,
"width": 1594
},
"sizes": {
"large": {
"h": 1548,
"w": 1594
}
},
"type": "photo",
"url": "https://t.co/R6aWzXD0yg"
}
]
},
"card": null,
"place": {},
"entities": {
"hashtags": [],
"symbols": [],
"urls": [
{
"display_url": "github.com/Blaizzy/mlx-vlm",
"expanded_url": "https://github.com/Blaizzy/mlx-vlm",
"indices": [
452,
475
],
"url": "https://t.co/7BvnEuzKvj"
}
],
"user_mentions": [
{
"id_str": "4783690002",
"indices": [
76,
91
],
"name": "Google DeepMind",
"screen_name": "GoogleDeepMind"
},
{
"id_str": "1175057008678055936",
"indices": [
129,
136
],
"name": "Technology Innovation Institute",
"screen_name": "TIIuae"
},
{
"id_str": "16319797",
"indices": [
162,
174
],
"name": "IBM Research",
"screen_name": "IBMResearch"
},
{
"id_str": "1174426031010152448",
"indices": [
225,
234
],
"name": "Facebook",
"screen_name": "facebook"
},
{
"id_str": "1144354342532722689",
"indices": [
274,
283
],
"name": "Roboflow",
"screen_name": "roboflow"
}
]
},
"quoted_tweet": null,
"retweeted_tweet": null,
"isLimitedReply": false,
"communityInfo": null,
"article": null
},
"retweeted_tweet": null,
"isLimitedReply": false,
"communityInfo": null,
"article": null
}