[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-qwen-qwen-mm-plugins-api":3,"mdc--rgsql1-key":37,"related-repo-qwen-qwen-mm-plugins-api":1043,"related-org-qwen-qwen-mm-plugins-api":1147},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":12,"stars":26,"repoUrl":27,"updatedAt":28,"license":29,"forks":30,"topics":31,"repo":32,"sourceUrl":35,"mdContent":36},"qwen-mm-plugins-api","analyze media with Qwen vision models","Cloud MCP tools for understanding media, by model family. VL model: vision_chat (caption\u002FVQA), ocr, grounding (detect\u002Flocate objects). Omni model (reads frames + audio together): timestamped captioning, ASR (plain \u002F controllable \u002F multi-speaker diarized), temporal grounding, event counting, music captioning. Plus transcribe_audio (ASR) and segmentation (SAM3). Use when a question about an image\u002Fvideo\u002Faudio needs an external model, not just local reading.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},"qwen","Qwen","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fqwen.png","QwenLM",[13,17,20,23],{"name":14,"slug":15,"type":16},"AI","ai","tag",{"name":18,"slug":19,"type":16},"MCP","mcp",{"name":21,"slug":22,"type":16},"Multimodal","multimodal",{"name":24,"slug":25,"type":16},"Computer Vision","computer-vision",54,"https:\u002F\u002Fgithub.com\u002FQwenLM\u002FQwen-MM-Plugins","2026-08-12T04:49:59.354141",null,0,[],{"repoUrl":27,"stars":26,"forks":30,"topics":33,"description":34},[],"Make any agent harness multimodal-native.","https:\u002F\u002Fgithub.com\u002FQwenLM\u002FQwen-MM-Plugins\u002Ftree\u002FHEAD\u002Fsrc\u002Fcapabilities\u002Fapi\u002Fskill","---\nname: qwen-mm-plugins-api\ndescription: \"Cloud MCP tools for understanding media, by model family. VL model: vision_chat (caption\u002FVQA), ocr, grounding (detect\u002Flocate objects). Omni model (reads frames + audio together): timestamped captioning, ASR (plain \u002F controllable \u002F multi-speaker diarized), temporal grounding, event counting, music captioning. Plus transcribe_audio (ASR) and segmentation (SAM3). Use when a question about an image\u002Fvideo\u002Faudio needs an external model, not just local reading.\"\n---\n\n# Qwen-MM-Plugins API\n\nYou have `qwen-mm-plugins-api` MCP tools available. They call external models\u002Fservices to understand media, grouped by model family:\n\n- **VL model** (Qwen-VL, OpenAI-compatible endpoint): `vision_chat`, `ocr`, `grounding`.\n- **Omni model** (Qwen-Omni — reads video frames **and** the embedded audio track together, so one call reasons over both): `omni_asr`, `omni_asr_timestamped`, `omni_multi_speaker_asr`, `omni_av_caption`, `omni_av_grounding`, `omni_av_counting`, `omni_music_caption`.\n- **Other services**: `transcribe_audio` (Qwen3-ASR), `segmentation` (a SAM3 server).\n\nPrefer these over manual ffmpeg\u002Fffprobe scripting. Check the `qwen-mm-plugins-api` tools in your tool list for full schemas and parameters.\n\n## When to Use Which Tool\n\n**VL model** (single images\u002Fvideos, spatial reasoning):\n\n- **Ask a VLM** about images\u002Fvideos (caption, VQA, free-form) → `vision_chat`\n- **Extract text** from an image → `ocr`\n- **Detect\u002Flocate objects** in an image (bounding boxes, spatial WHERE) → `grounding`\n\n**Omni model** (audio + video together, temporal reasoning; clips up to a few minutes):\n\n- **Transcribe speech, plain text** → `omni_asr` (one continuous string, no timestamps)\n- **Transcribe with timestamps** → `omni_asr_timestamped` (`granularity` = `sentence` or `word`; also returns SRT)\n- **Who said what** → `omni_multi_speaker_asr` (diarization: speaker labels + timestamps + SRT; pass `num_speakers` if known)\n- **Describe the content over time** → `omni_av_caption` (splits into spans, one description + start\u002Fend per span)\n- **Find WHEN something happens** → `omni_av_grounding` (natural-language `query` → matching time segments; temporal localization)\n- **Count how many times** an event\u002Fobject\u002Faction occurs → `omni_av_counting` (`target` → total + per-occurrence timestamps)\n- **Analyze \u002F caption a music track** → `omni_music_caption` (whole-track tags — genre \u002F moods \u002F instruments \u002F key \u002F time signature \u002F vocal profile — plus a dense English caption for music generation; audio-only, no timestamps)\n\n**Other services**:\n\n- **Segment objects** in an image (masks) → `segmentation`\n- **Transcribe speech** from audio\u002Fvideo, fast and long-file friendly → `transcribe_audio`\n\n## Tips\n\n**Vision chat**: pass `images`\u002F`videos` + `text` prompt. Default model `qwen3.7-plus`. Use `dry_run=true` to inspect payloads. Details in `references\u002Fvision_chat.md`.\n\n**Grounding**: returns normalized boxes (0–1000). Set `return_img=true` to get the annotated image back, or draw them yourself with core's `draw_bbox`. Needs `DASHSCOPE_API_KEY`.\n\n**ASR** (`transcribe_audio`): accepts audio or video, auto-chunks long files. Formats: `srt` (default), `text`, `json`. Needs `DASHSCOPE_API_KEY` (and `ffmpeg` to pull the audio track from a video).\n\n**Segmentation**: needs a SAM3 server (`SAM3_SERVER_URL`). To stand one up, run `references\u002Flaunch_sam3_server.py` (multi-GPU HTTP server; see its header for prerequisites).\n\n**Omni tools**: every tool takes a local audio\u002Fvideo `file_path` (or an http\u002FOSS URL) and supports `dry_run=true`. The AV tools (`caption`\u002F`grounding`\u002F`counting`) accept `fps` and `max_pixels` to trade temporal\u002Fspatial detail against token cost — raise `fps` only for fast\u002Ffrequent events; keep `max_pixels` at the default (≈448²) unless fine detail matters. The ASR family sends only the (extracted) audio track, so it is cheaper on video input. Timestamps are seconds from the start. Pass `language` (e.g. `zh`, `en`) as a hint when known. Default model `qwen3.5-omni-plus`; override per call with `model`.\n\n**Video delivery (VL & Omni)**: a local video is uploaded and sampled server-side (lifting the inline frame cap) when OSS is configured (`OSS_AK`\u002F`OSS_SK`\u002F`OSS_ENDPOINT`\u002F`OSS_BUCKET` + the `oss` extra); otherwise it is sampled into inline frames. Server-side sampling has a per-model video-duration limit (e.g. qwen3.7-plus 2 h, Qwen3.5-Omni 1 h), so a local file longer than that skips the upload and degrades to local frame sampling (VL: frames; Omni: frames + audio) — sparse for very long clips, but it still returns a result.\n\n## Choosing between the families (do NOT overlap)\n\n- **`transcribe_audio` vs `omni_asr*`**: `transcribe_audio` uses the dedicated Qwen3-ASR service (fast, chunks long files, 27 languages) — cheapest for a straight, long-file transcription. Pick the `omni_asr*` tools when you want Omni's understanding: multi-speaker diarization, controllable word\u002Fsentence granularity, or transcription fused with visual context.\n- **`grounding` (spatial, WHERE) vs `omni_av_grounding` (temporal, WHEN)**: `grounding` draws a bounding box in a single image; `omni_av_grounding` locates a span in time. Different axes — don't substitute one for the other.\n- **`vision_chat` vs the Omni AV tools**: `vision_chat` is a general VLM over images\u002Fvideo frames (no audio); the Omni tools fuse frames with the audio track and return structured, timestamped output. Use Omni when audio or precise timing matters.\n\n## Relationship to Other Capabilities (do NOT overlap)\n\n- **Read\u002Fvisualize local files** (images, video frames, PDF, Office, 3D, ...) → `qwen-mm-plugins-core` (`read_image`\u002F`read_video`\u002F`visualize`\u002F`crop`\u002F`draw_bbox`\u002F`save_view`).\n- **Confirm a fact or identify an entity** (reverse image \u002F web) → `qwen-mm-plugins-search` (`image_search`\u002F`web_search`\u002F`web_extractor`).\n- **Long videos (30 min+)**: for whole-video QA over long content, use the `qwen-mm-plugins-video-memory` skill (hierarchical graph memory) instead of feeding the entire file to these per-call tools.\n",{"data":38,"body":39},{"name":4,"description":6},{"type":40,"children":41},"root",[42,50,64,195,207,214,223,271,280,444,453,486,492,548,581,633,659,773,820,826,919,925],{"type":43,"tag":44,"props":45,"children":46},"element","h1",{"id":4},[47],{"type":48,"value":49},"text","Qwen-MM-Plugins API",{"type":43,"tag":51,"props":52,"children":53},"p",{},[54,56,62],{"type":48,"value":55},"You have ",{"type":43,"tag":57,"props":58,"children":60},"code",{"className":59},[],[61],{"type":48,"value":4},{"type":48,"value":63}," MCP tools available. They call external models\u002Fservices to understand media, grouped by model family:",{"type":43,"tag":65,"props":66,"children":67},"ul",{},[68,103,169],{"type":43,"tag":69,"props":70,"children":71},"li",{},[72,78,80,86,88,94,95,101],{"type":43,"tag":73,"props":74,"children":75},"strong",{},[76],{"type":48,"value":77},"VL model",{"type":48,"value":79}," (Qwen-VL, OpenAI-compatible endpoint): ",{"type":43,"tag":57,"props":81,"children":83},{"className":82},[],[84],{"type":48,"value":85},"vision_chat",{"type":48,"value":87},", ",{"type":43,"tag":57,"props":89,"children":91},{"className":90},[],[92],{"type":48,"value":93},"ocr",{"type":48,"value":87},{"type":43,"tag":57,"props":96,"children":98},{"className":97},[],[99],{"type":48,"value":100},"grounding",{"type":48,"value":102},".",{"type":43,"tag":69,"props":104,"children":105},{},[106,111,113,118,120,126,127,133,134,140,141,147,148,154,155,161,162,168],{"type":43,"tag":73,"props":107,"children":108},{},[109],{"type":48,"value":110},"Omni model",{"type":48,"value":112}," (Qwen-Omni — reads video frames ",{"type":43,"tag":73,"props":114,"children":115},{},[116],{"type":48,"value":117},"and",{"type":48,"value":119}," the embedded audio track together, so one call reasons over both): ",{"type":43,"tag":57,"props":121,"children":123},{"className":122},[],[124],{"type":48,"value":125},"omni_asr",{"type":48,"value":87},{"type":43,"tag":57,"props":128,"children":130},{"className":129},[],[131],{"type":48,"value":132},"omni_asr_timestamped",{"type":48,"value":87},{"type":43,"tag":57,"props":135,"children":137},{"className":136},[],[138],{"type":48,"value":139},"omni_multi_speaker_asr",{"type":48,"value":87},{"type":43,"tag":57,"props":142,"children":144},{"className":143},[],[145],{"type":48,"value":146},"omni_av_caption",{"type":48,"value":87},{"type":43,"tag":57,"props":149,"children":151},{"className":150},[],[152],{"type":48,"value":153},"omni_av_grounding",{"type":48,"value":87},{"type":43,"tag":57,"props":156,"children":158},{"className":157},[],[159],{"type":48,"value":160},"omni_av_counting",{"type":48,"value":87},{"type":43,"tag":57,"props":163,"children":165},{"className":164},[],[166],{"type":48,"value":167},"omni_music_caption",{"type":48,"value":102},{"type":43,"tag":69,"props":170,"children":171},{},[172,177,179,185,187,193],{"type":43,"tag":73,"props":173,"children":174},{},[175],{"type":48,"value":176},"Other services",{"type":48,"value":178},": ",{"type":43,"tag":57,"props":180,"children":182},{"className":181},[],[183],{"type":48,"value":184},"transcribe_audio",{"type":48,"value":186}," (Qwen3-ASR), ",{"type":43,"tag":57,"props":188,"children":190},{"className":189},[],[191],{"type":48,"value":192},"segmentation",{"type":48,"value":194}," (a SAM3 server).",{"type":43,"tag":51,"props":196,"children":197},{},[198,200,205],{"type":48,"value":199},"Prefer these over manual ffmpeg\u002Fffprobe scripting. Check the ",{"type":43,"tag":57,"props":201,"children":203},{"className":202},[],[204],{"type":48,"value":4},{"type":48,"value":206}," tools in your tool list for full schemas and parameters.",{"type":43,"tag":208,"props":209,"children":211},"h2",{"id":210},"when-to-use-which-tool",[212],{"type":48,"value":213},"When to Use Which Tool",{"type":43,"tag":51,"props":215,"children":216},{},[217,221],{"type":43,"tag":73,"props":218,"children":219},{},[220],{"type":48,"value":77},{"type":48,"value":222}," (single images\u002Fvideos, spatial reasoning):",{"type":43,"tag":65,"props":224,"children":225},{},[226,241,256],{"type":43,"tag":69,"props":227,"children":228},{},[229,234,236],{"type":43,"tag":73,"props":230,"children":231},{},[232],{"type":48,"value":233},"Ask a VLM",{"type":48,"value":235}," about images\u002Fvideos (caption, VQA, free-form) → ",{"type":43,"tag":57,"props":237,"children":239},{"className":238},[],[240],{"type":48,"value":85},{"type":43,"tag":69,"props":242,"children":243},{},[244,249,251],{"type":43,"tag":73,"props":245,"children":246},{},[247],{"type":48,"value":248},"Extract text",{"type":48,"value":250}," from an image → ",{"type":43,"tag":57,"props":252,"children":254},{"className":253},[],[255],{"type":48,"value":93},{"type":43,"tag":69,"props":257,"children":258},{},[259,264,266],{"type":43,"tag":73,"props":260,"children":261},{},[262],{"type":48,"value":263},"Detect\u002Flocate objects",{"type":48,"value":265}," in an image (bounding boxes, spatial WHERE) → ",{"type":43,"tag":57,"props":267,"children":269},{"className":268},[],[270],{"type":48,"value":100},{"type":43,"tag":51,"props":272,"children":273},{},[274,278],{"type":43,"tag":73,"props":275,"children":276},{},[277],{"type":48,"value":110},{"type":48,"value":279}," (audio + video together, temporal reasoning; clips up to a few minutes):",{"type":43,"tag":65,"props":281,"children":282},{},[283,300,340,364,380,404,428],{"type":43,"tag":69,"props":284,"children":285},{},[286,291,293,298],{"type":43,"tag":73,"props":287,"children":288},{},[289],{"type":48,"value":290},"Transcribe speech, plain text",{"type":48,"value":292}," → ",{"type":43,"tag":57,"props":294,"children":296},{"className":295},[],[297],{"type":48,"value":125},{"type":48,"value":299}," (one continuous string, no timestamps)",{"type":43,"tag":69,"props":301,"children":302},{},[303,308,309,314,316,322,324,330,332,338],{"type":43,"tag":73,"props":304,"children":305},{},[306],{"type":48,"value":307},"Transcribe with timestamps",{"type":48,"value":292},{"type":43,"tag":57,"props":310,"children":312},{"className":311},[],[313],{"type":48,"value":132},{"type":48,"value":315}," (",{"type":43,"tag":57,"props":317,"children":319},{"className":318},[],[320],{"type":48,"value":321},"granularity",{"type":48,"value":323}," = ",{"type":43,"tag":57,"props":325,"children":327},{"className":326},[],[328],{"type":48,"value":329},"sentence",{"type":48,"value":331}," or ",{"type":43,"tag":57,"props":333,"children":335},{"className":334},[],[336],{"type":48,"value":337},"word",{"type":48,"value":339},"; also returns SRT)",{"type":43,"tag":69,"props":341,"children":342},{},[343,348,349,354,356,362],{"type":43,"tag":73,"props":344,"children":345},{},[346],{"type":48,"value":347},"Who said what",{"type":48,"value":292},{"type":43,"tag":57,"props":350,"children":352},{"className":351},[],[353],{"type":48,"value":139},{"type":48,"value":355}," (diarization: speaker labels + timestamps + SRT; pass ",{"type":43,"tag":57,"props":357,"children":359},{"className":358},[],[360],{"type":48,"value":361},"num_speakers",{"type":48,"value":363}," if known)",{"type":43,"tag":69,"props":365,"children":366},{},[367,372,373,378],{"type":43,"tag":73,"props":368,"children":369},{},[370],{"type":48,"value":371},"Describe the content over time",{"type":48,"value":292},{"type":43,"tag":57,"props":374,"children":376},{"className":375},[],[377],{"type":48,"value":146},{"type":48,"value":379}," (splits into spans, one description + start\u002Fend per span)",{"type":43,"tag":69,"props":381,"children":382},{},[383,388,389,394,396,402],{"type":43,"tag":73,"props":384,"children":385},{},[386],{"type":48,"value":387},"Find WHEN something happens",{"type":48,"value":292},{"type":43,"tag":57,"props":390,"children":392},{"className":391},[],[393],{"type":48,"value":153},{"type":48,"value":395}," (natural-language ",{"type":43,"tag":57,"props":397,"children":399},{"className":398},[],[400],{"type":48,"value":401},"query",{"type":48,"value":403}," → matching time segments; temporal localization)",{"type":43,"tag":69,"props":405,"children":406},{},[407,412,414,419,420,426],{"type":43,"tag":73,"props":408,"children":409},{},[410],{"type":48,"value":411},"Count how many times",{"type":48,"value":413}," an event\u002Fobject\u002Faction occurs → ",{"type":43,"tag":57,"props":415,"children":417},{"className":416},[],[418],{"type":48,"value":160},{"type":48,"value":315},{"type":43,"tag":57,"props":421,"children":423},{"className":422},[],[424],{"type":48,"value":425},"target",{"type":48,"value":427}," → total + per-occurrence timestamps)",{"type":43,"tag":69,"props":429,"children":430},{},[431,436,437,442],{"type":43,"tag":73,"props":432,"children":433},{},[434],{"type":48,"value":435},"Analyze \u002F caption a music track",{"type":48,"value":292},{"type":43,"tag":57,"props":438,"children":440},{"className":439},[],[441],{"type":48,"value":167},{"type":48,"value":443}," (whole-track tags — genre \u002F moods \u002F instruments \u002F key \u002F time signature \u002F vocal profile — plus a dense English caption for music generation; audio-only, no timestamps)",{"type":43,"tag":51,"props":445,"children":446},{},[447,451],{"type":43,"tag":73,"props":448,"children":449},{},[450],{"type":48,"value":176},{"type":48,"value":452},":",{"type":43,"tag":65,"props":454,"children":455},{},[456,471],{"type":43,"tag":69,"props":457,"children":458},{},[459,464,466],{"type":43,"tag":73,"props":460,"children":461},{},[462],{"type":48,"value":463},"Segment objects",{"type":48,"value":465}," in an image (masks) → ",{"type":43,"tag":57,"props":467,"children":469},{"className":468},[],[470],{"type":48,"value":192},{"type":43,"tag":69,"props":472,"children":473},{},[474,479,481],{"type":43,"tag":73,"props":475,"children":476},{},[477],{"type":48,"value":478},"Transcribe speech",{"type":48,"value":480}," from audio\u002Fvideo, fast and long-file friendly → ",{"type":43,"tag":57,"props":482,"children":484},{"className":483},[],[485],{"type":48,"value":184},{"type":43,"tag":208,"props":487,"children":489},{"id":488},"tips",[490],{"type":48,"value":491},"Tips",{"type":43,"tag":51,"props":493,"children":494},{},[495,500,502,508,510,516,518,523,525,531,533,539,541,547],{"type":43,"tag":73,"props":496,"children":497},{},[498],{"type":48,"value":499},"Vision chat",{"type":48,"value":501},": pass ",{"type":43,"tag":57,"props":503,"children":505},{"className":504},[],[506],{"type":48,"value":507},"images",{"type":48,"value":509},"\u002F",{"type":43,"tag":57,"props":511,"children":513},{"className":512},[],[514],{"type":48,"value":515},"videos",{"type":48,"value":517}," + ",{"type":43,"tag":57,"props":519,"children":521},{"className":520},[],[522],{"type":48,"value":48},{"type":48,"value":524}," prompt. Default model ",{"type":43,"tag":57,"props":526,"children":528},{"className":527},[],[529],{"type":48,"value":530},"qwen3.7-plus",{"type":48,"value":532},". Use ",{"type":43,"tag":57,"props":534,"children":536},{"className":535},[],[537],{"type":48,"value":538},"dry_run=true",{"type":48,"value":540}," to inspect payloads. Details in ",{"type":43,"tag":57,"props":542,"children":544},{"className":543},[],[545],{"type":48,"value":546},"references\u002Fvision_chat.md",{"type":48,"value":102},{"type":43,"tag":51,"props":549,"children":550},{},[551,556,558,564,566,572,574,580],{"type":43,"tag":73,"props":552,"children":553},{},[554],{"type":48,"value":555},"Grounding",{"type":48,"value":557},": returns normalized boxes (0–1000). Set ",{"type":43,"tag":57,"props":559,"children":561},{"className":560},[],[562],{"type":48,"value":563},"return_img=true",{"type":48,"value":565}," to get the annotated image back, or draw them yourself with core's ",{"type":43,"tag":57,"props":567,"children":569},{"className":568},[],[570],{"type":48,"value":571},"draw_bbox",{"type":48,"value":573},". Needs ",{"type":43,"tag":57,"props":575,"children":577},{"className":576},[],[578],{"type":48,"value":579},"DASHSCOPE_API_KEY",{"type":48,"value":102},{"type":43,"tag":51,"props":582,"children":583},{},[584,589,590,595,597,603,605,610,611,617,618,623,625,631],{"type":43,"tag":73,"props":585,"children":586},{},[587],{"type":48,"value":588},"ASR",{"type":48,"value":315},{"type":43,"tag":57,"props":591,"children":593},{"className":592},[],[594],{"type":48,"value":184},{"type":48,"value":596},"): accepts audio or video, auto-chunks long files. Formats: ",{"type":43,"tag":57,"props":598,"children":600},{"className":599},[],[601],{"type":48,"value":602},"srt",{"type":48,"value":604}," (default), ",{"type":43,"tag":57,"props":606,"children":608},{"className":607},[],[609],{"type":48,"value":48},{"type":48,"value":87},{"type":43,"tag":57,"props":612,"children":614},{"className":613},[],[615],{"type":48,"value":616},"json",{"type":48,"value":573},{"type":43,"tag":57,"props":619,"children":621},{"className":620},[],[622],{"type":48,"value":579},{"type":48,"value":624}," (and ",{"type":43,"tag":57,"props":626,"children":628},{"className":627},[],[629],{"type":48,"value":630},"ffmpeg",{"type":48,"value":632}," to pull the audio track from a video).",{"type":43,"tag":51,"props":634,"children":635},{},[636,641,643,649,651,657],{"type":43,"tag":73,"props":637,"children":638},{},[639],{"type":48,"value":640},"Segmentation",{"type":48,"value":642},": needs a SAM3 server (",{"type":43,"tag":57,"props":644,"children":646},{"className":645},[],[647],{"type":48,"value":648},"SAM3_SERVER_URL",{"type":48,"value":650},"). To stand one up, run ",{"type":43,"tag":57,"props":652,"children":654},{"className":653},[],[655],{"type":48,"value":656},"references\u002Flaunch_sam3_server.py",{"type":48,"value":658}," (multi-GPU HTTP server; see its header for prerequisites).",{"type":43,"tag":51,"props":660,"children":661},{},[662,667,669,675,677,682,684,690,691,696,697,703,705,711,713,719,721,726,728,733,735,741,743,749,750,756,758,764,766,772],{"type":43,"tag":73,"props":663,"children":664},{},[665],{"type":48,"value":666},"Omni tools",{"type":48,"value":668},": every tool takes a local audio\u002Fvideo ",{"type":43,"tag":57,"props":670,"children":672},{"className":671},[],[673],{"type":48,"value":674},"file_path",{"type":48,"value":676}," (or an http\u002FOSS URL) and supports ",{"type":43,"tag":57,"props":678,"children":680},{"className":679},[],[681],{"type":48,"value":538},{"type":48,"value":683},". The AV tools (",{"type":43,"tag":57,"props":685,"children":687},{"className":686},[],[688],{"type":48,"value":689},"caption",{"type":48,"value":509},{"type":43,"tag":57,"props":692,"children":694},{"className":693},[],[695],{"type":48,"value":100},{"type":48,"value":509},{"type":43,"tag":57,"props":698,"children":700},{"className":699},[],[701],{"type":48,"value":702},"counting",{"type":48,"value":704},") accept ",{"type":43,"tag":57,"props":706,"children":708},{"className":707},[],[709],{"type":48,"value":710},"fps",{"type":48,"value":712}," and ",{"type":43,"tag":57,"props":714,"children":716},{"className":715},[],[717],{"type":48,"value":718},"max_pixels",{"type":48,"value":720}," to trade temporal\u002Fspatial detail against token cost — raise ",{"type":43,"tag":57,"props":722,"children":724},{"className":723},[],[725],{"type":48,"value":710},{"type":48,"value":727}," only for fast\u002Ffrequent events; keep ",{"type":43,"tag":57,"props":729,"children":731},{"className":730},[],[732],{"type":48,"value":718},{"type":48,"value":734}," at the default (≈448²) unless fine detail matters. The ASR family sends only the (extracted) audio track, so it is cheaper on video input. Timestamps are seconds from the start. Pass ",{"type":43,"tag":57,"props":736,"children":738},{"className":737},[],[739],{"type":48,"value":740},"language",{"type":48,"value":742}," (e.g. ",{"type":43,"tag":57,"props":744,"children":746},{"className":745},[],[747],{"type":48,"value":748},"zh",{"type":48,"value":87},{"type":43,"tag":57,"props":751,"children":753},{"className":752},[],[754],{"type":48,"value":755},"en",{"type":48,"value":757},") as a hint when known. Default model ",{"type":43,"tag":57,"props":759,"children":761},{"className":760},[],[762],{"type":48,"value":763},"qwen3.5-omni-plus",{"type":48,"value":765},"; override per call with ",{"type":43,"tag":57,"props":767,"children":769},{"className":768},[],[770],{"type":48,"value":771},"model",{"type":48,"value":102},{"type":43,"tag":51,"props":774,"children":775},{},[776,781,783,789,790,796,797,803,804,810,812,818],{"type":43,"tag":73,"props":777,"children":778},{},[779],{"type":48,"value":780},"Video delivery (VL & Omni)",{"type":48,"value":782},": a local video is uploaded and sampled server-side (lifting the inline frame cap) when OSS is configured (",{"type":43,"tag":57,"props":784,"children":786},{"className":785},[],[787],{"type":48,"value":788},"OSS_AK",{"type":48,"value":509},{"type":43,"tag":57,"props":791,"children":793},{"className":792},[],[794],{"type":48,"value":795},"OSS_SK",{"type":48,"value":509},{"type":43,"tag":57,"props":798,"children":800},{"className":799},[],[801],{"type":48,"value":802},"OSS_ENDPOINT",{"type":48,"value":509},{"type":43,"tag":57,"props":805,"children":807},{"className":806},[],[808],{"type":48,"value":809},"OSS_BUCKET",{"type":48,"value":811}," + the ",{"type":43,"tag":57,"props":813,"children":815},{"className":814},[],[816],{"type":48,"value":817},"oss",{"type":48,"value":819}," extra); otherwise it is sampled into inline frames. Server-side sampling has a per-model video-duration limit (e.g. qwen3.7-plus 2 h, Qwen3.5-Omni 1 h), so a local file longer than that skips the upload and degrades to local frame sampling (VL: frames; Omni: frames + audio) — sparse for very long clips, but it still returns a result.",{"type":43,"tag":208,"props":821,"children":823},{"id":822},"choosing-between-the-families-do-not-overlap",[824],{"type":48,"value":825},"Choosing between the families (do NOT overlap)",{"type":43,"tag":65,"props":827,"children":828},{},[829,863,898],{"type":43,"tag":69,"props":830,"children":831},{},[832,848,849,854,856,861],{"type":43,"tag":73,"props":833,"children":834},{},[835,840,842],{"type":43,"tag":57,"props":836,"children":838},{"className":837},[],[839],{"type":48,"value":184},{"type":48,"value":841}," vs ",{"type":43,"tag":57,"props":843,"children":845},{"className":844},[],[846],{"type":48,"value":847},"omni_asr*",{"type":48,"value":178},{"type":43,"tag":57,"props":850,"children":852},{"className":851},[],[853],{"type":48,"value":184},{"type":48,"value":855}," uses the dedicated Qwen3-ASR service (fast, chunks long files, 27 languages) — cheapest for a straight, long-file transcription. Pick the ",{"type":43,"tag":57,"props":857,"children":859},{"className":858},[],[860],{"type":48,"value":847},{"type":48,"value":862}," tools when you want Omni's understanding: multi-speaker diarization, controllable word\u002Fsentence granularity, or transcription fused with visual context.",{"type":43,"tag":69,"props":864,"children":865},{},[866,883,884,889,891,896],{"type":43,"tag":73,"props":867,"children":868},{},[869,874,876,881],{"type":43,"tag":57,"props":870,"children":872},{"className":871},[],[873],{"type":48,"value":100},{"type":48,"value":875}," (spatial, WHERE) vs ",{"type":43,"tag":57,"props":877,"children":879},{"className":878},[],[880],{"type":48,"value":153},{"type":48,"value":882}," (temporal, WHEN)",{"type":48,"value":178},{"type":43,"tag":57,"props":885,"children":887},{"className":886},[],[888],{"type":48,"value":100},{"type":48,"value":890}," draws a bounding box in a single image; ",{"type":43,"tag":57,"props":892,"children":894},{"className":893},[],[895],{"type":48,"value":153},{"type":48,"value":897}," locates a span in time. Different axes — don't substitute one for the other.",{"type":43,"tag":69,"props":899,"children":900},{},[901,911,912,917],{"type":43,"tag":73,"props":902,"children":903},{},[904,909],{"type":43,"tag":57,"props":905,"children":907},{"className":906},[],[908],{"type":48,"value":85},{"type":48,"value":910}," vs the Omni AV tools",{"type":48,"value":178},{"type":43,"tag":57,"props":913,"children":915},{"className":914},[],[916],{"type":48,"value":85},{"type":48,"value":918}," is a general VLM over images\u002Fvideo frames (no audio); the Omni tools fuse frames with the audio track and return structured, timestamped output. Use Omni when audio or precise timing matters.",{"type":43,"tag":208,"props":920,"children":922},{"id":921},"relationship-to-other-capabilities-do-not-overlap",[923],{"type":48,"value":924},"Relationship to Other Capabilities (do NOT overlap)",{"type":43,"tag":65,"props":926,"children":927},{},[928,987,1025],{"type":43,"tag":69,"props":929,"children":930},{},[931,936,938,944,945,951,952,958,959,965,966,972,973,978,979,985],{"type":43,"tag":73,"props":932,"children":933},{},[934],{"type":48,"value":935},"Read\u002Fvisualize local files",{"type":48,"value":937}," (images, video frames, PDF, Office, 3D, ...) → ",{"type":43,"tag":57,"props":939,"children":941},{"className":940},[],[942],{"type":48,"value":943},"qwen-mm-plugins-core",{"type":48,"value":315},{"type":43,"tag":57,"props":946,"children":948},{"className":947},[],[949],{"type":48,"value":950},"read_image",{"type":48,"value":509},{"type":43,"tag":57,"props":953,"children":955},{"className":954},[],[956],{"type":48,"value":957},"read_video",{"type":48,"value":509},{"type":43,"tag":57,"props":960,"children":962},{"className":961},[],[963],{"type":48,"value":964},"visualize",{"type":48,"value":509},{"type":43,"tag":57,"props":967,"children":969},{"className":968},[],[970],{"type":48,"value":971},"crop",{"type":48,"value":509},{"type":43,"tag":57,"props":974,"children":976},{"className":975},[],[977],{"type":48,"value":571},{"type":48,"value":509},{"type":43,"tag":57,"props":980,"children":982},{"className":981},[],[983],{"type":48,"value":984},"save_view",{"type":48,"value":986},").",{"type":43,"tag":69,"props":988,"children":989},{},[990,995,997,1003,1004,1010,1011,1017,1018,1024],{"type":43,"tag":73,"props":991,"children":992},{},[993],{"type":48,"value":994},"Confirm a fact or identify an entity",{"type":48,"value":996}," (reverse image \u002F web) → ",{"type":43,"tag":57,"props":998,"children":1000},{"className":999},[],[1001],{"type":48,"value":1002},"qwen-mm-plugins-search",{"type":48,"value":315},{"type":43,"tag":57,"props":1005,"children":1007},{"className":1006},[],[1008],{"type":48,"value":1009},"image_search",{"type":48,"value":509},{"type":43,"tag":57,"props":1012,"children":1014},{"className":1013},[],[1015],{"type":48,"value":1016},"web_search",{"type":48,"value":509},{"type":43,"tag":57,"props":1019,"children":1021},{"className":1020},[],[1022],{"type":48,"value":1023},"web_extractor",{"type":48,"value":986},{"type":43,"tag":69,"props":1026,"children":1027},{},[1028,1033,1035,1041],{"type":43,"tag":73,"props":1029,"children":1030},{},[1031],{"type":48,"value":1032},"Long videos (30 min+)",{"type":48,"value":1034},": for whole-video QA over long content, use the ",{"type":43,"tag":57,"props":1036,"children":1038},{"className":1037},[],[1039],{"type":48,"value":1040},"qwen-mm-plugins-video-memory",{"type":48,"value":1042}," skill (hierarchical graph memory) instead of feeding the entire file to these per-call tools.",{"items":1044,"total":1146},[1045,1052,1071,1085,1104,1119,1135],{"slug":4,"name":4,"fn":5,"description":6,"org":1046,"tags":1047,"stars":26,"repoUrl":27,"updatedAt":28},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1048,1049,1050,1051],{"name":14,"slug":15,"type":16},{"name":24,"slug":25,"type":16},{"name":18,"slug":19,"type":16},{"name":21,"slug":22,"type":16},{"slug":1053,"name":1053,"fn":1054,"description":1055,"org":1056,"tags":1057,"stars":26,"repoUrl":27,"updatedAt":1070},"qwen-mm-plugins-blender","build and edit 3D scenes in Blender","Use whenever a task involves building or editing a 3D scene or asset in Blender — modeling, characters\u002Fpeople, architecture\u002Finteriors, terrain\u002Flandscapes, props, materials, lighting, or rendering. Covers discovering installed add-ons, using generators, importing and REFINING ready-made assets, and matching the result to the spec. Requires a running Blender instance with the blender-mcp addon (see Prerequisite).",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1058,1061,1064,1067],{"name":1059,"slug":1060,"type":16},"3D","3d",{"name":1062,"slug":1063,"type":16},"Animation","animation",{"name":1065,"slug":1066,"type":16},"Creative","creative",{"name":1068,"slug":1069,"type":16},"Design","design","2026-08-05T05:58:30.425245",{"slug":943,"name":943,"fn":1072,"description":1073,"org":1074,"tags":1075,"stars":26,"repoUrl":27,"updatedAt":1084},"analyze and visualize multimodal files","Local MCP tools to read and visualize any file — images, video, documents, code, data, 3D, notebooks, and more — plus image tools for cropping, annotating, and extracting frames.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1076,1079,1082,1083],{"name":1077,"slug":1078,"type":16},"Data Analysis","data-analysis",{"name":1080,"slug":1081,"type":16},"LLM","llm",{"name":18,"slug":19,"type":16},{"name":21,"slug":22,"type":16},"2026-08-12T04:25:59.187593",{"slug":1086,"name":1086,"fn":1087,"description":1088,"org":1089,"tags":1090,"stars":26,"repoUrl":27,"updatedAt":1103},"qwen-mm-plugins-edu-agent","generate math problem solving tutorial videos","Generate step-by-step math problem-solving tutorial videos in Chinese (Mandarin).\nUse when: (1) a user provides a math problem and wants an explanation video,\n(2) someone says \"make a math tutorial\", \"explain this equation\", \"create a\nteaching video for this problem\", \"讲解这道题\", \"生成解题视频\",\n(3) the user wants a Chinese-language math lesson covering formulas, equations,\nor geometric figures, (4) the user shares a math problem in text or LaTeX and\nasks for a video walkthrough, (5) the input is an image_assets\u002F folder\ncontaining problem images — the skill will extract the problem via visual\nrecognition, solve it, and generate a tutorial video. Teaching components are\nrendered as realistic objects (solid opaque panels, 3D cards, SVG figures) with a\nmodern aurora mesh aesthetic.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1091,1094,1097,1100],{"name":1092,"slug":1093,"type":16},"Content Creation","content-creation",{"name":1095,"slug":1096,"type":16},"Education","education",{"name":1098,"slug":1099,"type":16},"Mathematics","mathematics",{"name":1101,"slug":1102,"type":16},"Video","video","2026-08-05T05:58:40.128581",{"slug":1105,"name":1105,"fn":1106,"description":1107,"org":1108,"tags":1109,"stars":26,"repoUrl":27,"updatedAt":1118},"qwen-mm-plugins-freecad","model parts and assemblies in FreeCAD","Use whenever a task involves parametric CAD in FreeCAD — modeling parts and assemblies, editing object properties, technical drawings, importing\u002Fexporting STEP\u002FSTL\u002FOBJ\u002FDXF, PDF\u002FExcel reports from a model, or finite-element (FEM\u002FCalculiX) analysis. Requires a running FreeCAD instance with the FreeCADMCP addon (see Prerequisite).",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1110,1111,1114,1115],{"name":1059,"slug":1060,"type":16},{"name":1112,"slug":1113,"type":16},"CAD","cad",{"name":1068,"slug":1069,"type":16},{"name":1116,"slug":1117,"type":16},"Engineering","engineering","2026-08-05T05:58:40.567406",{"slug":1002,"name":1002,"fn":1120,"description":1121,"org":1122,"tags":1123,"stars":26,"repoUrl":27,"updatedAt":1134},"search and extract web and image data","Web and reverse-image search MCP tools (Serper) for confirming facts — web_search (find facts), web_extractor (read a page in depth), image_search (reverse-search a frame to identify an entity). Use to verify anything you cannot confirm from the media alone.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1124,1127,1128,1131],{"name":1125,"slug":1126,"type":16},"Data Extraction","data-extraction",{"name":18,"slug":19,"type":16},{"name":1129,"slug":1130,"type":16},"Search","search",{"name":1132,"slug":1133,"type":16},"Web Scraping","web-scraping","2026-08-12T04:50:09.275491",{"slug":1040,"name":1040,"fn":1136,"description":1137,"org":1138,"tags":1139,"stars":26,"repoUrl":27,"updatedAt":1145},"analyze long form video content","Triggered for long videos (30+ minutes), whether a single file or a directory of multiple videos. Vision-language MCP tools designed for efficient reading and semantic analysis of long videos (30+ minutes), supporting memory construction and semantic search.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1140,1143,1144],{"name":1141,"slug":1142,"type":16},"Analysis","analysis",{"name":21,"slug":22,"type":16},{"name":1101,"slug":1102,"type":16},"2026-08-05T05:58:28.007509",7,{"items":1148,"total":1327},[1149,1164,1181,1197,1213,1227,1241,1255,1268,1282,1294,1309],{"slug":1150,"name":1150,"fn":1151,"description":1152,"org":1153,"tags":1154,"stars":1161,"repoUrl":1162,"updatedAt":1163},"batch","execute parallel batch operations on files","Execute batch operations on multiple files in parallel. Automatically discovers files, splits into chunks, and processes with parallel worker agents. Use `\u002Fbatch` followed by operation and file pattern.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1155,1158],{"name":1156,"slug":1157,"type":16},"Automation","automation",{"name":1159,"slug":1160,"type":16},"Concurrency","concurrency",26008,"https:\u002F\u002Fgithub.com\u002FQwenLM\u002Fqwen-code","2026-07-20T05:58:45.650067",{"slug":1165,"name":1165,"fn":1166,"description":1167,"org":1168,"tags":1169,"stars":1161,"repoUrl":1162,"updatedAt":1180},"cua-driver","drive native GUI applications via MCP","Drive a native GUI app (macOS, Windows, Linux) via the Qwen Cua Driver CLI (default) or MCP server; snapshot its accessibility tree, act through snapshot-bound element tokens, native menu paths, exact window geometry, or pixel coordinates, and verify from fresh state. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real application on the host.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1170,1173,1176,1179],{"name":1171,"slug":1172,"type":16},"Accessibility","accessibility",{"name":1174,"slug":1175,"type":16},"Browser Automation","browser-automation",{"name":1177,"slug":1178,"type":16},"Desktop","desktop",{"name":18,"slug":19,"type":16},"2026-08-09T04:04:41.832404",{"slug":1182,"name":1182,"fn":1183,"description":1184,"org":1185,"tags":1186,"stars":1161,"repoUrl":1162,"updatedAt":1196},"dataviz","design and validate data visualizations","Design guidance for charts, graphs, dashboards, maps, and data visualizations, including a local palette validator.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1187,1190,1193],{"name":1188,"slug":1189,"type":16},"Charts","charts",{"name":1191,"slug":1192,"type":16},"Dashboards","dashboards",{"name":1194,"slug":1195,"type":16},"Data Visualization","data-visualization","2026-07-16T05:59:28.31755",{"slug":1198,"name":1198,"fn":1199,"description":1200,"org":1201,"tags":1202,"stars":1161,"repoUrl":1162,"updatedAt":1212},"extension-creator","create Qwen Code extensions","Create, scaffold, customize, validate, and locally test Qwen Code extensions. Use when the user wants a new Qwen Code extension, needs help choosing an extension template, wants to add QWEN.md context, commands, skills, agents, MCP servers, settings, hooks, channels, or LSP servers, or asks how to link and test an extension locally. Invoke with `\u002Fextension-creator` followed by an extension path and optional template name.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1203,1206,1209],{"name":1204,"slug":1205,"type":16},"Coding","coding",{"name":1207,"slug":1208,"type":16},"Documentation","documentation",{"name":1210,"slug":1211,"type":16},"Plugin Development","plugin-development","2026-07-16T05:59:24.818967",{"slug":1214,"name":1214,"fn":1215,"description":1216,"org":1217,"tags":1218,"stars":1161,"repoUrl":1162,"updatedAt":1226},"loop","run scheduled or self-paced prompt loops","Create a loop that runs a prompt now and follows up either on a fixed schedule or through self-paced wakeups. Usage - \u002Floop check the build, \u002Floop 5m check the build, \u002Floop check the PR every 30m. \u002Floop list to show jobs, \u002Floop clear to cancel all.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1219,1220,1223],{"name":1156,"slug":1157,"type":16},{"name":1221,"slug":1222,"type":16},"Productivity","productivity",{"name":1224,"slug":1225,"type":16},"Scheduling","scheduling","2026-07-16T05:59:25.50027",{"slug":1228,"name":1228,"fn":1229,"description":1230,"org":1231,"tags":1232,"stars":1161,"repoUrl":1162,"updatedAt":1240},"new-app","build new applications from scratch","Workflow for creating new applications from scratch. Covers requirements gathering, tech stack selection, scaffolding, implementation, and delivery of a functional prototype.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1233,1234,1237],{"name":1116,"slug":1117,"type":16},{"name":1235,"slug":1236,"type":16},"Prototyping","prototyping",{"name":1238,"slug":1239,"type":16},"Web Development","web-development","2026-07-16T05:59:25.157573",{"slug":1242,"name":1242,"fn":1243,"description":1244,"org":1245,"tags":1246,"stars":1161,"repoUrl":1162,"updatedAt":1254},"qc-helper","provide Qwen Code usage and configuration support","Answer any question about Qwen Code usage, features, configuration, and troubleshooting by referencing the official user documentation. Also helps users view or modify their settings.json. Invoke with `\u002Fqc-helper` followed by a question, e.g. `\u002Fqc-helper how do I configure MCP servers?` or `\u002Fqc-helper change approval mode to yolo`.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1247,1250,1251],{"name":1248,"slug":1249,"type":16},"Configuration","configuration",{"name":1207,"slug":1208,"type":16},{"name":1252,"slug":1253,"type":16},"Reference","reference","2026-07-16T05:59:29.118413",{"slug":1256,"name":1256,"fn":1257,"description":1258,"org":1259,"tags":1260,"stars":1161,"repoUrl":1162,"updatedAt":1267},"review","review code for quality and security","Review changed code for correctness, security, code quality, and performance. Use when the user asks to review code changes, a PR, or specific files. Invoke with `\u002Freview`, `\u002Freview \u003Cpr-number>`, `\u002Freview \u003Cfile-path>`, `\u002Freview \u003Cpr-number> --comment` to post inline comments on the PR, or `\u002Freview --fix` to apply the findings to your working tree. Add `--effort low|medium|high` to trade depth for speed (defaults to high for PRs, medium for local changes).",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1261,1264],{"name":1262,"slug":1263,"type":16},"Code Review","code-review",{"name":1265,"slug":1266,"type":16},"Security","security","2026-08-12T04:27:21.296311",{"slug":1269,"name":1269,"fn":1270,"description":1271,"org":1272,"tags":1273,"stars":1161,"repoUrl":1162,"updatedAt":1281},"simplify","clean up and simplify code changes","Review recent code changes for reuse, code quality, and efficiency, then directly apply straightforward cleanup improvements. Use when the user wants a post-implementation cleanup pass, pre-PR polish, or asks to simplify\u002Frefine recent changes. Invoke with `\u002Fsimplify` or `\u002Fsimplify \u003Cfocus>`.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1274,1277,1278],{"name":1275,"slug":1276,"type":16},"Code Analysis","code-analysis",{"name":1116,"slug":1117,"type":16},{"name":1279,"slug":1280,"type":16},"Performance","performance","2026-07-20T05:58:46.653811",{"slug":1283,"name":1283,"fn":1284,"description":1285,"org":1286,"tags":1287,"stars":1161,"repoUrl":1162,"updatedAt":1293},"stuck","diagnose and debug stuck Qwen Code sessions","Diagnose frozen, stuck, or slow Qwen Code sessions on this machine. Scans for problematic processes, high CPU\u002Fmemory usage, hung subprocesses, and debug logs. Use \u002Fstuck or \u002Fstuck \u003CPID> to focus on a specific process.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1288,1291,1292],{"name":1289,"slug":1290,"type":16},"Debugging","debugging",{"name":1116,"slug":1117,"type":16},{"name":1279,"slug":1280,"type":16},"2026-07-16T05:59:25.838629",{"slug":1295,"name":1295,"fn":1296,"description":1297,"org":1298,"tags":1299,"stars":1306,"repoUrl":1307,"updatedAt":1308},"open-computer-use","configure Open Computer Use MCP server","Platform-neutral guidance for using Open Computer Use, the open-source Computer Use MCP server and CLI for macOS, Linux, and Windows. Use when an agent needs to install, verify, troubleshoot, configure, or operate Open Computer Use through its native CLI, stdio MCP server, or direct Computer Use tool calls.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1300,1301,1304,1305],{"name":1156,"slug":1157,"type":16},{"name":1302,"slug":1303,"type":16},"CLI","cli",{"name":1177,"slug":1178,"type":16},{"name":18,"slug":19,"type":16},176,"https:\u002F\u002Fgithub.com\u002FQwenLM\u002Fopen-computer-use","2026-07-16T05:59:22.010839",{"slug":1310,"name":1310,"fn":1311,"description":1312,"org":1313,"tags":1314,"stars":1324,"repoUrl":1325,"updatedAt":1326},"auto-pr","automate pull request submission and review","Automated PR submission assistant, including code review, documentation generation, and PR creation",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1315,1316,1317,1318,1321],{"name":1156,"slug":1157,"type":16},{"name":1262,"slug":1263,"type":16},{"name":1207,"slug":1208,"type":16},{"name":1319,"slug":1320,"type":16},"GitHub","github",{"name":1322,"slug":1323,"type":16},"Pull Requests","pull-requests",136,"https:\u002F\u002Fgithub.com\u002FQwenLM\u002Fqwen-code-examples","2026-07-16T06:00:01.890524",28]