@vllm_project
🚀 Great to see vLLM powering OCR at this scale — Chandra-OCR-2 (5B) serving ~60 papers/hour per L40S across 16 parallel jobs. The full pipeline breakdown is a great read 👇 🔗 https://t.co/z8tkv04ZLp https://t.co/vi9FPj6JIQ
🐦 Twitter Post Details
Viewing enriched Twitter post
📊 Media Metadata
{
"media": [
{
"url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2043964594679636260/media_0.jpg",
"media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2043964594679636260/media_0.jpg",
"type": "photo",
"filename": "media_0.jpg"
},
{
"url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2043964594679636260/media_1.jpg",
"media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2043964594679636260/media_1.jpg",
"type": "photo",
"filename": "media_1.jpg"
}
],
"processed_at": "2026-04-14T14:50:43.518850",
"pipeline_version": "2.0"
}🔧 Raw API Response
{
"type": "tweet",
"id": "2043964594679636260",
"url": "https://x.com/vllm_project/status/2043964594679636260",
"twitterUrl": "https://twitter.com/vllm_project/status/2043964594679636260",
"text": "🚀 Great to see vLLM powering OCR at this scale — Chandra-OCR-2 (5B) serving ~60 papers/hour per L40S across 16 parallel jobs.\n\nThe full pipeline breakdown is a great read 👇\n🔗 https://t.co/z8tkv04ZLp https://t.co/vi9FPj6JIQ",
"source": "Twitter for iPhone",
"retweetCount": 19,
"replyCount": 4,
"likeCount": 128,
"quoteCount": 1,
"viewCount": 9214,
"createdAt": "Tue Apr 14 08:08:29 +0000 2026",
"lang": "en",
"bookmarkCount": 44,
"isReply": false,
"inReplyToId": null,
"conversationId": "2043964594679636260",
"displayTextRange": [
0,
198
],
"inReplyToUserId": null,
"inReplyToUsername": null,
"author": {
"type": "user",
"userName": "vllm_project",
"url": "https://x.com/vllm_project",
"twitterUrl": "https://twitter.com/vllm_project",
"id": "1774187564276289536",
"name": "vLLM",
"isVerified": false,
"isBlueVerified": true,
"verifiedType": null,
"profilePicture": "https://pbs.twimg.com/profile_images/1774187681746182144/N_5NJ8B1_normal.jpg",
"coverPicture": "https://pbs.twimg.com/profile_banners/1774187564276289536/1733806062",
"description": "",
"location": "",
"followers": 35100,
"following": 33,
"status": "",
"canDm": true,
"canMediaTag": true,
"createdAt": "Sat Mar 30 21:31:01 +0000 2024",
"entities": {
"description": {
"urls": []
},
"url": {}
},
"fastFollowersCount": 0,
"favouritesCount": 558,
"hasCustomTimelines": true,
"isTranslator": false,
"mediaCount": 250,
"statusesCount": 906,
"withheldInCountries": [],
"affiliatesHighlightedLabel": {},
"possiblySensitive": false,
"pinnedTweetIds": [],
"profile_bio": {
"description": "A high-throughput and memory-efficient inference and serving engine for LLMs. Join https://t.co/lxJ0SfX5pJ to discuss together with the community!",
"entities": {
"description": {
"hashtags": [],
"symbols": [],
"urls": [
{
"display_url": "slack.vllm.ai",
"expanded_url": "http://slack.vllm.ai",
"indices": [
83,
106
],
"url": "https://t.co/lxJ0SfX5pJ"
}
],
"user_mentions": []
},
"url": {
"urls": [
{
"display_url": "vllm.ai",
"expanded_url": "https://vllm.ai",
"indices": [
0,
23
],
"url": "https://t.co/OSHR5C6pCV"
}
]
}
}
},
"isAutomated": false,
"automatedBy": null
},
"extendedEntities": {
"media": [
{
"display_url": "pic.twitter.com/vi9FPj6JIQ",
"expanded_url": "https://twitter.com/vllm_project/status/2043964594679636260/photo/1",
"ext_media_availability": {
"status": "Available"
},
"features": {
"large": {
"faces": []
},
"orig": {
"faces": []
}
},
"id_str": "2043963658385141762",
"indices": [
199,
222
],
"media_key": "3_2043963658385141762",
"media_results": {
"id": "QXBpTWVkaWFSZXN1bHRzOgwAAQoAARxdnh4v2rACCgACHF2e+C9a4SQAAA==",
"result": {
"__typename": "ApiMedia",
"id": "QXBpTWVkaWE6DAABCgABHF2eHi/asAIKAAIcXZ74L1rhJAAA",
"media_key": "3_2043963658385141762"
}
},
"media_url_https": "https://pbs.twimg.com/media/HF2eHi_asAINFSs.jpg",
"original_info": {
"focus_rects": [
{
"h": 1038,
"w": 1854,
"x": 0,
"y": 0
},
{
"h": 1502,
"w": 1502,
"x": 0,
"y": 0
},
{
"h": 1502,
"w": 1318,
"x": 0,
"y": 0
},
{
"h": 1502,
"w": 751,
"x": 227,
"y": 0
},
{
"h": 1502,
"w": 1854,
"x": 0,
"y": 0
}
],
"height": 1502,
"width": 1854
},
"sizes": {
"large": {
"h": 1502,
"w": 1854
}
},
"type": "photo",
"url": "https://t.co/vi9FPj6JIQ"
}
]
},
"card": null,
"place": {},
"entities": {
"hashtags": [],
"symbols": [],
"timestamps": [],
"urls": [
{
"display_url": "huggingface.co/blog/nielsr/oc…",
"expanded_url": "https://huggingface.co/blog/nielsr/ocr-papers-jobs",
"indices": [
175,
198
],
"url": "https://t.co/z8tkv04ZLp"
}
],
"user_mentions": []
},
"quoted_tweet": {
"type": "tweet",
"id": "2043779449322160270",
"url": "https://x.com/ClementDelangue/status/2043779449322160270",
"twitterUrl": "https://twitter.com/ClementDelangue/status/2043779449322160270",
"text": "We just OCR'd 27,000 arxiv papers into Markdown using an open 5B model, 16 parallel HF Jobs on L40S GPUs, and a mounted bucket.\n\nTotal cost: $850 Total time: ~29 hours Jobs that crashed: 0\n\nThis now powers \"Chat with your paper\" on https://t.co/G2mDae0uv9 https://t.co/qpz7Q9x8Od",
"source": "Twitter for iPhone",
"retweetCount": 209,
"replyCount": 75,
"likeCount": 1952,
"quoteCount": 21,
"viewCount": 121657,
"createdAt": "Mon Apr 13 19:52:46 +0000 2026",
"lang": "en",
"bookmarkCount": 1375,
"isReply": false,
"inReplyToId": null,
"conversationId": "2043779449322160270",
"displayTextRange": [
0,
255
],
"inReplyToUserId": null,
"inReplyToUsername": null,
"author": {
"type": "user",
"userName": "ClementDelangue",
"url": "https://x.com/ClementDelangue",
"twitterUrl": "https://twitter.com/ClementDelangue",
"id": "186420551",
"name": "clem 🤗",
"isVerified": false,
"isBlueVerified": true,
"verifiedType": null,
"profilePicture": "https://pbs.twimg.com/profile_images/1100512198139498497/utHSJ4st_normal.png",
"coverPicture": "https://pbs.twimg.com/profile_banners/186420551/1731168567",
"description": "",
"location": "Miami / New York / Paris",
"followers": 248929,
"following": 4836,
"status": "",
"canDm": true,
"canMediaTag": true,
"createdAt": "Fri Sep 03 12:25:12 +0000 2010",
"entities": {
"description": {
"urls": []
},
"url": {}
},
"fastFollowersCount": 0,
"favouritesCount": 8641,
"hasCustomTimelines": true,
"isTranslator": false,
"mediaCount": 2822,
"statusesCount": 29444,
"withheldInCountries": [],
"affiliatesHighlightedLabel": {},
"possiblySensitive": false,
"pinnedTweetIds": [
"2039695447506210905"
],
"profile_bio": {
"description": "Co-founder & CEO @HuggingFace 🤗, the open and collaborative platform for AI builders",
"entities": {
"description": {
"hashtags": [],
"symbols": [],
"urls": [],
"user_mentions": [
{
"id_str": "0",
"indices": [
17,
29
],
"name": "",
"screen_name": "HuggingFace"
}
]
},
"url": {
"urls": [
{
"display_url": "huggingface.co/clem",
"expanded_url": "http://huggingface.co/clem",
"indices": [
0,
23
],
"url": "https://t.co/7OMlqfI2Rh"
}
]
}
}
},
"isAutomated": false,
"automatedBy": null
},
"extendedEntities": {
"media": [
{
"display_url": "pic.twitter.com/qpz7Q9x8Od",
"expanded_url": "https://twitter.com/ClementDelangue/status/2043779449322160270/photo/1",
"ext_media_availability": {
"status": "Available"
},
"features": {
"large": {
"faces": []
},
"orig": {
"faces": []
}
},
"id_str": "2043779385258381312",
"indices": [
256,
279
],
"media_key": "3_2043779385258381312",
"media_results": {
"id": "QXBpTWVkaWFSZXN1bHRzOgwAAQoAARxc9oXBl5AACgACHFz2lKwXQI4AAA==",
"result": {
"__typename": "ApiMedia",
"id": "QXBpTWVkaWE6DAABCgABHFz2hcGXkAAKAAIcXPaUrBdAjgAA",
"media_key": "3_2043779385258381312"
}
},
"media_url_https": "https://pbs.twimg.com/media/HFz2hcGXkAAWWMU.jpg",
"original_info": {
"focus_rects": [
{
"h": 784,
"w": 1400,
"x": 0,
"y": 116
},
{
"h": 900,
"w": 900,
"x": 144,
"y": 0
},
{
"h": 900,
"w": 789,
"x": 200,
"y": 0
},
{
"h": 900,
"w": 450,
"x": 369,
"y": 0
},
{
"h": 900,
"w": 1400,
"x": 0,
"y": 0
}
],
"height": 900,
"width": 1400
},
"sizes": {
"large": {
"h": 900,
"w": 1400
}
},
"type": "photo",
"url": "https://t.co/qpz7Q9x8Od"
}
]
},
"card": null,
"place": {},
"entities": {
"hashtags": [],
"symbols": [],
"timestamps": [],
"urls": [
{
"display_url": "hf.co/papers",
"expanded_url": "http://hf.co/papers",
"indices": [
232,
255
],
"url": "https://t.co/G2mDae0uv9"
}
],
"user_mentions": []
},
"quoted_tweet": null,
"retweeted_tweet": null,
"isLimitedReply": false,
"communityInfo": null,
"article": null
},
"retweeted_tweet": null,
"isLimitedReply": false,
"communityInfo": null,
"article": null
}