{"slug": "qwen3-vl", "name": "Qwen3-VL", "category": "ai", "category_name": "AI models and assistants", "description": "", "recommenders": 3, "commenters": 6, "warners": 0, "score": 3.355, "rising": 0.35, "first": "2025-11-30", "latest": "2026-08-06", "why": "i use qwen-30b-vl all the time locally as my goto model because it's just so insanely fast", "why_is_quote": true, "has_verdict": false, "sentiment": {"recommend": 3, "neutral": 4, "compare": 0, "warn": 0}, "spark": [1, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0], "listed": true, "loved": [], "caveat": "", "short_quote": "i use qwen-30b-vl all the time locally as my goto model because it's just so insanely fast", "icon": true, "domain": null, "github_repo": null, "official_url": "https://github.com/QwenLM/Qwen3-VL", "pricing": "", "aliases": ["Qwen3-VL"], "verdict": null, "praise": [{"label": "", "summary": "", "quotes": [{"id": 787267217, "item_id": 49204201, "url": "https://news.ycombinator.com/item?id=49204201", "by": "vunderba", "date": "2026-08-06", "stance": "recommend", "quote": "Qwen3-VL picks up new images in a NAS, auto captions and adds the text descriptions as a hidden EXIF layer", "excerpt": "Oh I do. Headless 128GB RAM machine serving llama.cpp with a number of local models that I use on a daily basis. \u2022 Qwen3-VL picks up new images in a NAS, auto captions and adds the text descriptions as a hidden EXIF layer into the image, which is used for fast search and organization in conjunction with a Qdrant vector database. \u2022 Gemma3:27b is used for personal translation work (mostly English and Chinese). \u2022 Some s", "context": "", "use_case": "auto-captioning images", "switched_from": "", "story": "Qwen3.8 Max now ranked as the best overall model by agentic index"}, {"id": 750646352, "item_id": 46915397, "url": "https://news.ycombinator.com/item?id=46915397", "by": "josefcub", "date": "2026-02-06", "stance": "recommend", "quote": "I generally run Qwen3-vl at 235b, at a Q4_K_M quantization level so that it fits", "excerpt": "I am! I moved from a shoebox Linux workstation with 32MB of RAM and a 12GB RTX 3060 to a 256GB M3 Ultra, mainly for unified memory. I've only had it a couple of months, but so far it's proving its worth in the quality of LLM output, even quantized. I generally run Qwen3-vl at 235b, at a Q4_K_M quantization level so that it fits, and it leaves me plenty of RAM for workstation tasks while delivering tokens at around 30", "context": "runs a 256GB M3 Ultra Mac Studio", "use_case": "local LLM on Mac Studio", "switched_from": "", "story": "Ask HN: Anyone Using a Mac Studio for Local AI/LLM?"}, {"id": 737522128, "item_id": 46095133, "url": "https://news.ycombinator.com/item?id=46095133", "by": "thot_experiment", "date": "2025-11-30", "stance": "recommend", "quote": "i use qwen-30b-vl all the time locally as my goto model because it's just so insanely fast", "excerpt": "anyone have a tl;dr for me on what the best way to get the video comprehension stuff going is? i use qwen-30b-vl all the time locally as my goto model because it's just so insanely fast, curious to mess with the video stuff, the vision comprehension works great and i use it for OCR and classification all the time", "context": "", "use_case": "local OCR and classification", "switched_from": "", "story": "Qwen3-VL can scan two-hour videos and pinpoint nearly every detail"}]}], "caveats": [], "who": [{"by": "ali_chherawalla", "context": "built an on-device Android AI app", "url": "https://news.ycombinator.com/item?id=47238566", "date": "2026-03-03"}, {"by": "josefcub", "context": "runs a 256GB M3 Ultra Mac Studio", "url": "https://news.ycombinator.com/item?id=46915397", "date": "2026-02-06"}], "tags": [], "use_cases": ["auto-captioning images", "on-device vision", "analyzing films for training data", "local llm on mac studio", "extracting narrative structure from films", "recognize dance moves in videos"], "compared": [{"slug": "qwen2-5", "name": "Qwen2.5", "co": 2, "switched_from": 0, "switched_to": 0}, {"slug": "llama-cpp", "name": "llama.cpp", "co": 2, "switched_from": 0, "switched_to": 0}], "makers": [], "threads": [{"url": "https://news.ycombinator.com/item?id=46094606", "title": "Qwen3-VL can scan two-hour videos and pinpoint nearly every detail", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=49200652", "title": "Qwen3.8 Max now ranked as the best overall model by agentic index", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=47238519", "title": "Show HN: Qwen 3.5 running on a $300 Android phone \u2013 on-device, open source", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=46933515", "title": "Show HN: Fine-tuned Qwen2.5-7B on 100 films for probabilistic story graphs", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=46907001", "title": "Ask HN: Anyone Using a Mac Studio for Local AI/LLM?", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=46628336", "title": "Show HN: Fine-tuned Qwen2.5-7B on 100 films for probabilistic story graphs", "mentions": 1}], "sentiment_by_month": {"2025-11": {"recommend": 1}, "2025-12": {"neutral": 1}, "2026-01": {"neutral": 1}, "2026-02": {"recommend": 1, "neutral": 1}, "2026-03": {"neutral": 1}, "2026-04": {}, "2026-05": {}, "2026-06": {}, "2026-07": {}, "2026-08": {"recommend": 1}, "2026-09": {}, "2026-10": {}}, "quotes": [{"id": 787267217, "item_id": 49204201, "url": "https://news.ycombinator.com/item?id=49204201", "by": "vunderba", "date": "2026-08-06", "stance": "recommend", "quote": "Qwen3-VL picks up new images in a NAS, auto captions and adds the text descriptions as a hidden EXIF layer", "excerpt": "Oh I do. Headless 128GB RAM machine serving llama.cpp with a number of local models that I use on a daily basis. \u2022 Qwen3-VL picks up new images in a NAS, auto captions and adds the text descriptions as a hidden EXIF layer into the image, which is used for fast search and organization in conjunction with a Qdrant vector database. \u2022 Gemma3:27b is used for personal translation work (mostly English and Chinese). \u2022 Some s", "context": "", "use_case": "auto-captioning images", "switched_from": "", "story": "Qwen3.8 Max now ranked as the best overall model by agentic index"}, {"id": 755817060, "item_id": 47238566, "url": "https://news.ycombinator.com/item?id=47238566", "by": "ali_chherawalla", "date": "2026-03-03", "stance": "neutral", "quote": "Vision uses SmolVLM and Qwen3-VL.", "excerpt": "ofcourse ofcourse, I've documented everything here: https://github.com/alichherawalla/off-grid-mobile-ai/blob/main/docs/ARCHITECTURE.md llama.cpp compiled as a native Android library via the NDK, linked into React Native through a custom JSI bridge. GGUF models loaded straight into memory. On Snapdragon devices we use QNN (Qualcomm Neural Network) for hardware acceleration. OpenCL GPU fallback on everything else. CPU", "context": "built an on-device Android AI app", "use_case": "on-device vision", "switched_from": "", "story": "Show HN: Qwen 3.5 running on a $300 Android phone \u2013 on-device, open source"}, {"id": 750936449, "item_id": 46933528, "url": "https://news.ycombinator.com/item?id=46933528", "by": "graphpilled", "date": "2026-02-08", "stance": "neutral", "quote": "I built a 1000+ line Python pipeline using Qwen3-VL to analyze each film with subtitles enabled.", "excerpt": "Hi HN, I'm a computer systems engineering student in Mexico who switched from film school. I built CineGraphs because my filmmaker friends and I kept hitting the same wall\u2014we'd have a vague idea for a film but no structured way to explore where it could go. Every AI writing tool we tried output generic, formulaic slop. I didn't want to build another ChatGPT wrapper, so I went a different route. The idea is simple: yo", "context": "", "use_case": "analyzing films for training data", "switched_from": "", "story": "Show HN: Fine-tuned Qwen2.5-7B on 100 films for probabilistic story graphs"}, {"id": 750646352, "item_id": 46915397, "url": "https://news.ycombinator.com/item?id=46915397", "by": "josefcub", "date": "2026-02-06", "stance": "recommend", "quote": "I generally run Qwen3-vl at 235b, at a Q4_K_M quantization level so that it fits", "excerpt": "I am! I moved from a shoebox Linux workstation with 32MB of RAM and a 12GB RTX 3060 to a 256GB M3 Ultra, mainly for unified memory. I've only had it a couple of months, but so far it's proving its worth in the quality of LLM output, even quantized. I generally run Qwen3-vl at 235b, at a Q4_K_M quantization level so that it fits, and it leaves me plenty of RAM for workstation tasks while delivering tokens at around 30", "context": "runs a 256GB M3 Ultra Mac Studio", "use_case": "local LLM on Mac Studio", "switched_from": "", "story": "Ask HN: Anyone Using a Mac Studio for Local AI/LLM?"}, {"id": 746059522, "item_id": 46628720, "url": "https://news.ycombinator.com/item?id=46628720", "by": "graphpilled", "date": "2026-01-14", "stance": "neutral", "quote": "I built a 1000+ line extraction pipeline using Qwen3-VL", "excerpt": "Hi HN, I'm a computer systems engineering student in Mexico who switched from film school. I built CineGraphs because my filmmaker friends and I kept hitting the same wall\u2014we'd have a vague idea for a film but no structured way to explore where it could go. Every AI writing tool we tried output generic, formulaic slop. I didn't want to build another ChatGPT wrapper, so I went a different route. The idea is simple: yo", "context": "", "use_case": "extracting narrative structure from films", "switched_from": "", "story": "Show HN: Fine-tuned Qwen2.5-7B on 100 films for probabilistic story graphs"}, {"id": 738049504, "item_id": 46128094, "url": "https://news.ycombinator.com/item?id=46128094", "by": "eurekin", "date": "2025-12-02", "stance": "neutral", "quote": "Insane if true... now I wonder, if I use it to go through some old dance routing video catalogue", "excerpt": "Insane if true... now I wonder, if I use it to go through some old dance routing video catalogue to recognize and write individual move lists", "context": "", "use_case": "recognize dance moves in videos", "switched_from": "", "story": "Qwen3-VL can scan two-hour videos and pinpoint nearly every detail"}, {"id": 737522128, "item_id": 46095133, "url": "https://news.ycombinator.com/item?id=46095133", "by": "thot_experiment", "date": "2025-11-30", "stance": "recommend", "quote": "i use qwen-30b-vl all the time locally as my goto model because it's just so insanely fast", "excerpt": "anyone have a tl;dr for me on what the best way to get the video comprehension stuff going is? i use qwen-30b-vl all the time locally as my goto model because it's just so insanely fast, curious to mess with the video stuff, the vision comprehension works great and i use it for OCR and classification all the time", "context": "", "use_case": "local OCR and classification", "switched_from": "", "story": "Qwen3-VL can scan two-hour videos and pinpoint nearly every detail"}]}