[{"_1":2,"_3":-5,"_4":-5},"loaderData",{"_5":6,"_7":8},"actionData","errors","root",{"_23":24,"_25":26,"_27":21,"_28":21,"_29":21,"_30":21,"_31":21,"_32":21,"_33":34},"routes/canvas.shared.$sharedTextdocId",{"_9":10,"_11":-7,"_12":13,"_14":15,"_16":17,"_18":19,"_20":21,"_22":21},"sharedTextdocId","68d3972d363881918f24524394a87d87","versionInt","title","Qwen Aug–sep 2025 Timeline (interactive)","type","code/react","content","import React, { useMemo, useState } from \"react\";\nimport { Calendar, Filter, Sparkles, Image, Mic, Languages, ShieldCheck, AudioLines, Workflow, Bot, Microscope } from \"lucide-react\";\n\n// --- Data ---------------------------------------------------------\n// Source: user-provided Qwen blog dump (Jan–Sep 2025)\nconst releases = [\n { date: \"2025-09-24\", title: \"Qwen3-Max\", tag: \"Release\", blurb: \"Largest and most capable Qwen model to date (preview ranks near top).\", icon: Sparkles },\n { date: \"2025-09-23\", title: \"Qwen3‑LiveTranslate‑Flash\", tag: \"Release\", blurb: \"Real-time multilingual A/V interpretation built on Qwen3‑Omni.\", icon: Languages },\n { date: \"2025-09-23\", title: \"Qwen3‑VL\", tag: \"Open-Source\", blurb: \"New flagship vision‑language series: sharper perception, deeper reasoning.\", icon: Image },\n { date: \"2025-09-22\", title: \"Travel Planner (Multi‑Agent)\", tag: \"Release\", blurb: \"Smart travel planning assistant with tool‑use + multi‑agent orchestration.\", icon: Workflow },\n { date: \"2025-09-22\", title: \"Qwen3Guard\", tag: \"Open-Source\", blurb: \"Real‑time safety guardrail model for token‑stream moderation.\", icon: ShieldCheck },\n { date: \"2025-09-21\", title: \"Qwen‑Image‑Edit‑2509\", tag: \"Open-Source\", blurb: \"Monthly image‑edit update: multi‑image support, better consistency.\", icon: Image },\n { date: \"2025-09-21\", title: \"Qwen3‑Omni\", tag: \"Open-Source\", blurb: \"Natively omni‑modal foundation model (text, image, audio, video).\", icon: Bot },\n { date: \"2025-09-10\", title: \"Qwen3‑TTS‑Flash\", tag: \"Open-Source\", blurb: \"Multi‑timbre, multi‑lingual, multi‑dialect expressive speech synthesis.\", icon: AudioLines },\n { date: \"2025-09-08\", title: \"Qwen3‑Next\", tag: \"Research\", blurb: \"Training/inference efficiency: context‑length & parameter scaling.\", icon: Microscope },\n { date: \"2025-08-18\", title: \"Qwen3‑ASR‑Flash\", tag: \"Research\", blurb: \"Speech recognition leveraging Qwen3‑Omni + large‑scale ASR data.\", icon: Mic },\n { date: \"2025-08-04\", title: \"Qwen‑Image‑Edit (first release)\", tag: \"Research\", blurb: \"Image‑editing variant of Qwen‑Image (20B) with strong text rendering.\", icon: Image },\n { date: \"2025-07-27\", title: \"Qwen‑Image (20B MMDiT)\", tag: \"Research\", blurb: \"Image foundation model with native text rendering & precise editing.\", icon: Image },\n { date: \"2025-07-24\", title: \"GSPO (Scalable RL for LMs)\", tag: \"Research\", blurb: \"RL scaling framework to enhance deep reasoning & problem‑solving.\", icon: Microscope },\n { date: \"2025-07-22\", title: \"Qwen3‑Coder‑480B‑Agentic\", tag: \"Release\", blurb: \"Most agentic code model; introduces powerful coding + tool‑use.\", icon: Bot },\n { date: \"2025-06-27\", title: \"Qwen‑TTS update (dialects)\", tag: \"Release\", blurb: \"Adds Chinese dialects + human‑level expressiveness via large‑scale data.\", icon: AudioLines },\n { date: \"2025-06-26\", title: \"Qwen VLo\", tag: \"Release\", blurb: \"Unified multimodal understanding + generation model.\", icon: Image },\n { date: \"2025-06-05\", title: \"Qwen3 Embedding\", tag: \"Open-Source\", blurb: \"Embedding + reranking models built on Qwen3 foundation.\", icon: Microscope },\n { date: \"2025-04-28\", title: \"Qwen3 (MoE + Dense open‑weights)\", tag: \"Open-Source\", blurb: \"Open‑weights for MoE (235B A22B; 30B A3B) and six dense models under Apache 2.0.\", icon: Sparkles },\n { date: \"2025-03-27\", title: \"QVQ‑Max (visual reasoning)\", tag: \"Open-Source\", blurb: \"First release of visual reasoning model.\", icon: Image },\n { date: \"2025-03-26\", title: \"Qwen2.5‑Omni\", tag: \"Open-Source\", blurb: \"End‑to‑end multimodal flagship (see, hear, talk, write, act).\", icon: Bot },\n { date: \"2025-03-23\", title: \"Qwen2.5‑VL‑32B‑Instruct\", tag: \"Open-Source\", blurb: \"Open‑sourced 32B VL model improved via RL; Apache 2.0.\", icon: Image },\n { date: \"2025-03-05\", title: \"QwQ‑32B\", tag: \"Release\", blurb: \"32B reasoning model achieving R1‑class results with far fewer params.\", icon: Microscope },\n { date: \"2025-02-24\", title: \"QwQ‑Max‑Preview\", tag: \"Release\", blurb: \"Preview of deep‑reasoning model for versatile problem‑solving.\", icon: Microscope },\n { date: \"2025-01-28\", title: \"Qwen2.5‑Max\", tag: \"Release\", blurb: \"Large‑scale MoE model (>20T tokens; SFT + RLHF).\", icon: Sparkles },\n { date: \"2025-01-26\", title: \"Qwen2.5‑1M (open‑source)\", tag: \"Open-Source\", blurb: \"Open‑source 1M‑token context models + inference framework.\", icon: Sparkles },\n { date: \"2025-01-26\", title: \"Qwen2.5‑VL\", tag: \"Open-Source\", blurb: \"New flagship VL model; major leap over Qwen2‑VL.\", icon: Image },\n { date: \"2025-01-20\", title: \"Global‑batch load balance (MoE training)\", tag: \"Research\", blurb: \"Technique to improve MoE LLM training via better load balancing.\", icon: Microscope },\n { date: \"2025-01-13\", title: \"Process Supervision PRM (math)\", tag: \"Release\", blurb: \"State‑of‑the‑art PRM for process supervision in mathematical reasoning.\", icon: Microscope },\n];\n\nconst TAGS = [\"All\", \"Release\", \"Open-Source\", \"Research\"] as const;\ntype Tag = typeof TAGS[number];\n\nconst Chip: React.FC<{ active?: boolean; onClick?: () => void; children: React.ReactNode }> = ({ active, onClick, children }) => (\n \n {children}\n \n);\n\nexport default function QwenTimeline() {\n const [filter, setFilter] = useState(\"All\");\n const [view, setView] = useState<\"timeline\"|\"bars\">(\"timeline\");\n\n const months = useMemo(() => {\n const labels = [\"Jan\",\"Feb\",\"Mar\",\"Apr\",\"May\",\"Jun\",\"Jul\",\"Aug\",\"Sep\"];\n return labels.map((m,i)=>({ key:`2025-${String(i+1).padStart(2,'0')}`, label:`${m} 2025`}));\n }, []);\n\n const byMonth = useMemo(()=>{\n const map: Record = {};\n for (const m of months) map[m.key]=[];\n for (const r of (filter===\"All\"?releases:releases.filter(x=>x.tag===filter))) {\n const k=r.date.slice(0,7);\n if(!map[k]) map[k]=[];\n map[k].push(r);\n }\n return map;\n },[filter,months]);\n\n return (\n

Qwen Releases — Jan → Sep 2025

\n \n {TAGS.map(t=>(setFilter(t)}>{t}))}\n

\n \n \n

\n\n

\n {view==='timeline'? (\n

\n {months.map(m=>(\n

{m.label}

\n {(byMonth[m.key]||[]).map((r,i)=>{\n const Icon=r.icon; const dd=new Date(r.date).toLocaleDateString(undefined,{day:'2-digit'});\n return (\n

\n \n {r.title}\n {dd}\n

\n );\n })}\n

\n ))}\n

\n ):(\n

\n {months.map(m=>{\n const items=byMonth[m.key]||[];\n return (\n

\n {items.map((r,i)=>{\n const color=r.tag==='Release'?'bg-blue-500/70':r.tag==='Open-Source'?'bg-emerald-500/70':'bg-purple-500/70';\n return

;\n })}\n

{items.length}

{m.label}

{r.title}

\n );\n })}\n

\n )}\n

\n );\n}\n","access","public","isModerationBlocked",false,"isAnonifyApiKeyDetected","dd",{"_35":36,"_37":38},"disablePrefetch",true,"shouldPrefetchAccount","shouldPrefetchUser","shouldPrefetchSystemHints","promoteCss","disableSSR","stripModulepreloadImports","statsigGateEvaluationsPromise",["P",34],"traceId","1050873605240877523","traceTime",1776568886079]