{"openapi":"3.1.0","info":{"title":"4chan API","version":"1.0.0","description":"Live read-only access to the 4chan imageboard, served straight from its official JSON API — one of the most influential anonymous forums on the internet and a primary source of internet culture, exposed as clean JSON with the HTML stripped to plain text. The boards endpoint lists every one of 4chan's ~77 boards with its code, title, worksafe flag, pages, threads-per-page and a description, from /g/ (Technology) and /v/ (Video Games) to /news/, /sci/ and /pol/. The catalog endpoint returns every live thread on a board summarized — the opening post's subject and a plain-text teaser of its comment, the reply and image counts, the sticky and closed flags and when it was last bumped — sorted by activity, replies or images: the front page of the board as data. The hot endpoint self-discovers and returns the single most active thread on a board right now, fully expanded with the original post and its top replies, so you never need a thread id that has expired. The thread endpoint returns one specific thread in full by its id — the original post and every reply with author name, timestamp, plain-text comment and image filename — and falls back to the current top thread when no id is given. Everything is the live board state, nothing stored. This is the imageboard layer for any social-listening, trend-tracking, meme-research, moderation or bot app. Live from 4chan's JSON API. Distinct from Reddit, Hacker News and Lemmy APIs — this is 4chan's boards, catalogs and threads. Some boards are NSFW (worksafe flag provided). 5 endpoints, no key on our side, nothing cached.","contact":{"name":"PremiumApi","url":"https://www.oanor.com/by/premiumapi"}},"servers":[{"url":"https://api.oanor.com/fourchan-api","description":"oanor gateway"}],"tags":[{"name":"Imageboard"},{"name":"Meta"}],"components":{"securitySchemes":{"oanorKey":{"type":"apiKey","in":"header","name":"x-oanor-key","description":"Get your key at https://www.oanor.com/developer/keys"}}},"security":[{"oanorKey":[]}],"paths":{"/v1/boards":{"get":{"operationId":"get_v1_boards","tags":["Imageboard"],"summary":"All boards","description":"","parameters":[{"name":"worksafe","in":"query","required":false,"description":"true to return only worksafe (SFW) boards","schema":{"type":"string"}}],"security":[{"oanorKey":[]}],"responses":{"200":{"description":"OK","content":{"application/json":{"example":{"data":{"count":77,"boards":[{"board":"3","pages":10,"title":"3DCG","worksafe":true,"bump_limit":310,"description":"\"/3/ - 3DCG\" is 4chan's board for 3D modeling and imagery.","image_limit":150,"threads_per_page":15},{"board":"a","pages":10,"title":"Anime & Manga","worksafe":true,"bump_limit":500,"description":"\"/a/ - Anime & Manga\" is 4chan's imageboard dedicated to the discussion of Japanese animation and manga.","image_limit":300,"threads_per_page":15},{"board":"aco","pages":10,"title":"Adult Cartoons","worksafe":false,"bump_limit":300,"description":"\"/aco/ - Adult Cartoons\" is 4chan's imageboard for posting western-styled adult cartoon artwork.","image_limit":300,"threads_per_page":15},{"board":"adv","pages":10,"title":"Advice","worksafe":true,"bump_limit":310,"description":"\"/adv/ - Advice\" is 4chan's board for giving and receiving advice.","image_limit":150,"threads_per_page":15},{"board":"an","pages":10,"title":"Animals & Nature","worksafe":true,"bump_limit":310,"description":"\"/an/ - Animals & Nature\" is 4chan's imageboard for posting pictures of animals, pets, and nature.","image_limit":150,"threads_per_page":15},{"board":"b","pages":10,"title":"Random","worksafe":false,"bump_limit":300,"description":"\"/b/ - Random\" is the birthplace of Anonymous, and where people go to discuss random topics and create memes on 4chan.","image_limit":150,"threads_per_page":15},{"board":"bant","pages":10,"title":"International/Random","worksafe":false,"bump_limit":300,"description":"\"/bant/ - International/Random\" is 4chan's international hanging out board, where you can have fun with Anonymous all over the world.","image_limit":150,"threads_per_page":15},{"board":"biz","pages":10,"title":"Business & Finance","worksafe":true,"bump_limit":310,"description":"\"/biz/ - Business & Finance\" is 4chan's imageboard for the discussion of business and finance, and cryptocurrencies such as Bitcoin and Dogecoin.","image_limit":150,"threads_per_page":20},{"board":"c","pages":10,"title":"Anime/Cute","worksafe":true,"bump_limit":310,"description":"\"/c/ - Anime/Cute\" is 4chan's imageboard for cute and moe anime images.","image_limit":150,"threads_per_page":15},{"board":"cgl","pages":10,"title":"Cosplay & EGL","worksafe":true,"bump_limit":310,"description":"\"/cgl/ - Cosplay & EGL\" is 4chan's imageboard for the discussion of cosplay, elegant gothic lolita (EGL), and anime conventions.","image_limit":150,"threads_per_page":15},{"board":"ck","pages":10,"title":"Food & Cooking","worksafe":true,"bump_limit":310,"description":"\"/ck/ - Food & Cooking\" is 4chan's imageboard for food pictures and cooking recipes.","image_limit":150,"threads_per_page":15},{"board":"cm","pages":10,"title":"Cute/Male","worksafe":true,"bump_limit":310,"description":"\"/cm/ - Cute/Male\" is 4chan's imageboard for posting pictures of cute anime males.","image_limit":150,"threads_per_page":15},{"board":"co","pages":10,"title":"Comics & Cartoons","worksafe":true,"bump_limit":500,"description":"\"/co/ - Comics & Cartoons\" is 4chan's imageboard dedicated to the discussion of Western cartoons and comics.","image_limit":300,"threads_per_page":15},{"board":"d","pages":10,"title":"Hentai/Alternative","worksafe":false,"bump_limit":300,"description":"\"/d/ - Hentai/Alternative\" is 4chan's imageboard for alternative hentai images.","image_limit":300,"threads_per_page":15},{"board":"diy","pages":10,"title":"Do It Yourself","worksafe":true,"bump_limit":310,"description":"\"/diy/ - Do It Yourself\" is 4chan's imageboard for DIY/do it yourself projects, home improvement, and makers.","image_limit":150,"threads_per_page":15},{"board":"e","pages":10,"title":"Ecchi","worksafe":false,"bump_limit":300,"description":"\"/e/ - Ecchi\" is 4chan's imageboard for suggestive (ecchi) hentai images.","image_limit":300,"threads_per_page":15},{"board":"f","pages":1,"title":"Flash","worksafe":false,"bump_limit":300,"description":"\"/f/ - Flash\" is 4chan's upload board for sharing Adobe Flash files (SWFs).","image_limit":300,"threads_per_page":30},{"board":"fa","pages":10,"title":"Fashion","worksafe":true,"bump_limit":310,"description":"\"/fa/ - Fashion\" is 4chan's imageboard for images and discussion relating to fashion and apparel.","image_limit":150,"threads_per_page":15},{"board":"fit","pages":10,"title":"Fitness","worksafe":true,"bump_limit":310,"description":"\"/fit/ - Fitness\" is 4chan's imageboard for weightlifting, health, and fitness.","image_limit":150,"threads_per_page":15},{"board":"g","pages":10,"title":"Technology","worksafe":true,"bump_limit":310,"description":"\"/g/ - Technology\" is 4chan's imageboard for discussing computer hardware and software, programming, and general technology.","image_limit":150,"threads_per_page":15},{"board":"gd","pages":10,"title":"Graphic Design","worksafe":true,"bump_limit":310,"description":"\"/gd/ - Graphic Design\" is 4chan's imageboard for graphic design.","image_limit":150,"threads_per_page":15},{"board":"gif","pages":5,"title":"Adult GIF","worksafe":false,"bump_limit":300,"description":"\"/gif/ - Adult GIF\" is 4chan's imageboard dedicated to animated adult GIFs and WEBMs.","image_limit":300,"threads_per_page":15},{"board":"h","pages":10,"title":"Hentai","worksafe":false,"bump_limit":300,"description":"\"/h/ - Hentai\" is 4chan's imageboard for adult Japanese anime hentai images.","image_limit":300,"threads_per_page":15},{"board":"hc","pages":10,"title":"Hardcore","worksafe":false,"bump_limit":300,"description":"\"/hc/ - Hardcore\" is 4chan's imageboard for the posting of adult hardcore pornography.","image_limit":300,"threads_per_page":15},{"board":"his","pages":10,"title":"History & Humanities","worksafe":true,"bump_limit":310,"description":"\"/his/ - History & Humanities\" is 4chan's board for discussing and debating history.","image_limit":150,"threads_per_page":15},{"board":"hm","pages":10,"title":"Handsome Men","worksafe":false,"bump_limit":300,"description":"\"/hm/ - Handsome Men\" is 4chan's imageboard dedicated to sharing adult images of handsome men.","image_limit":150,"threads_per_page":15},{"board":"hr","pages":10,"title":"High Resolution","worksafe":false,"bump_limit":300,"description":"\"/hr/ - High Resolution\" is 4chan's imageboard for the sharing of high resolution images.","image_limit":300,"threads_per_page":15},{"board":"i","pages":10,"title":"Oekaki","worksafe":false,"bump_limit":300,"description":"\"/i/ - Oekaki\" is 4chan's oekaki board for drawing and sharing art.","image_limit":300,"threads_per_page":15},{"board":"ic","pages":10,"title":"Artwork/Critique","worksafe":false,"bump_limit":300,"description":"\"/ic/ - Artwork/Critique\" is 4chan's imageboard for the discussion and critique of art.","image_limit":300,"threads_per_page":15},{"board":"int","pages":10,"title":"International","worksafe":true,"bump_limit":310,"description":"\"/int/ - International\" is 4chan's international board, for the exchange of foreign language and culture.","image_limit":150,"threads_per_page":15},{"board":"jp","pages":10,"title":"Otaku Culture","worksafe":true,"bump_limit":310,"description":"\"/jp/ - Otaku Culture\" is 4chan's board for discussing Japanese otaku culture.","image_limit":300,"threads_per_page":15},{"board":"k","pages":10,"title":"Weapons","worksafe":true,"bump_limit":310,"description":"\"/k/ - Weapons\" is 4chan's imageboard for discussing all types of weaponry, from military tanks to guns and knives.","image_limit":150,"threads_per_page":15},{"board":"lgbt","pages":10,"title":"LGBT","worksafe":true,"bump_limit":310,"description":"\"/lgbt/ - LGBT\" is 4chan's imageboard for Lesbian-Gay-Bisexual-Transgender-Queer and sexuality discussion.","image_limit":150,"threads_per_page":15},{"board":"lit","pages":10,"title":"Literature","worksafe":true,"bump_limit":310,"description":"\"/lit/ - Literature\" is 4chan's board for the discussion of books, authors, and literature.","image_limit":150,"threads_per_page":15},{"board":"m","pages":10,"title":"Mecha","worksafe":true,"bump_limit":310,"description":"\"/m/ - Mecha\" is 4chan's imageboard for discussing Japanese mecha robots and anime, like Gundam and Macross.","image_limit":150,"threads_per_page":15},{"board":"mlp","pages":10,"title":"Pony","worksafe":true,"bump_limit":500,"description":"\"/mlp/ - Pony\" is 4chan's imageboard dedicated to the discussion of My Little Pony: Friendship is Magic.","image_limit":300,"threads_per_page":15},{"board":"mu","pages":10,"title":"Music","worksafe":true,"bump_limit":310,"description":"\"/mu/ - Music\" is 4chan's imageboard for discussing all types of music.","image_limit":150,"threads_per_page":15},{"board":"n","pages":10,"title":"Transportation","worksafe":true,"bump_limit":310,"description":"\"/n/ - Transportation\" is 4chan's imageboard for discussing modes of transportation like trains and bicycles.","image_limit":150,"threads_per_page":15},{"board":"news","pages":10,"title":"Current News","worksafe":true,"bump_limit":500,"description":"\"/news/ - Current News; is 4chan's board for current news.","image_limit":300,"threads_per_page":15},{"board":"o","pages":10,"title":"Auto","worksafe":true,"bump_limit":310,"description":"\"/o/ - Auto\" is 4chan's imageboard for discussing cars and motorcycles.","image_limit":150,"threads_per_page":15},{"board":"out","pages":10,"title":"Outdoors","worksafe":true,"bump_limit":310,"description":"\"/out/ - Outdoors\" is 4chan's imageboard for discussing survivalist skills and outdoor activities such as hiking.","image_limit":150,"threads_per_page":15},{"board":"p","pages":10,"title":"Photography","worksafe":true,"bump_limit":310,"description":"\"/p/ - Photography\" is 4chan's imageboard for sharing and critiquing photos.","image_limit":150,"threads_per_page":15},{"board":"po","pages":10,"title":"Papercraft & Origami","worksafe":true,"bump_limit":310,"description":"\"/po/ - Papercraft & Origami\" is 4chan's imageboard for posting papercraft and origami templates and instructions.","image_limit":150,"threads_per_page":15},{"board":"pol","pages":10,"title":"Politically Incorrect","worksafe":false,"bump_limit":300,"description":"\"/pol/ - Politically Incorrect\" is 4chan's board for discussing and debating politics and current events.","image_limit":300,"threads_per_page":20},{"board":"pw","pages":10,"title":"Professional Wrestling","worksafe":true,"bump_limit":310,"description":"\"/pw/ - Professional Wrestling\" is 4chan's imageboard for the discussion of professional wrestling.","image_limit":150,"threads_per_page":15},{"board":"qst","pages":10,"title":"Quests","worksafe":true,"bump_limit":750,"description":"\"/qst/ - Quests\" is 4chan's imageboard for grinding XP.","image_limit":375,"threads_per_page":15},{"board":"r","pages":10,"title":"Adult Requests","worksafe":false,"bump_limit":300,"description":"\"/r/ - Request\" is 4chan's imageboard dedicated to fulfilling all types of user requests.","image_limit":300,"threads_per_page":15},{"board":"r9k","pages":10,"title":"ROBOT9001","worksafe":false,"bump_limit":500,"description":"\"/r9k/ - ROBOT9001\" is a board for hanging out and posting greentext stories.","image_limit":150,"threads_per_page":15},{"board":"s","pages":10,"title":"Sexy Beautiful Women","worksafe":false,"bump_limit":300,"description":"\"/s/ - Sexy Beautiful Women\" is 4chan's imageboard dedicated to sharing images of softcore pornography.","image_limit":150,"threads_per_page":15},{"board":"s4s","pages":10,"title":"Shit 4chan Says","worksafe":false,"bump_limit":300,"description":"\"[s4s] - Shit 4chan Says\" is 4chan's imageboard for posting dank memes :^)","image_limit":300,"threads_per_page":15},{"board":"sci","pages":10,"title":"Science & Math","worksafe":true,"bump_limit":310,"description":"\"/sci/ - Science & Math\" is 4chan's board for the discussion of science and math.","image_limit":250,"threads_per_page":15},{"board":"soc","pages":10,"title":"Cams & Meetups","worksafe":false,"bump_limit":500,"description":"\"/soc/ - Cams & Meetups\" is 4chan's board for camwhores and meetups.","image_limit":300,"threads_per_page":15},{"board":"sp","pages":10,"title":"Sports","worksafe":true,"bump_limit":500,"description":"\"/sp/ - Sports\" is 4chan's imageboard for sports discussion.","image_limit":300,"threads_per_page":15},{"board":"t","pages":10,"title":"Torrents","worksafe":false,"bump_limit":300,"description":"\"/t/ - Torrents\" is 4chan's imageboard for posting links and descriptions to torrents.","image_limit":300,"threads_per_page":15},{"board":"tg","pages":10,"title":"Traditional Games","worksafe":true,"bump_limit":310,"description":"\"/tg/ - Traditional Games\" is 4chan's imageboard for discussing traditional gaming, such as board games and tabletop RPGs.","image_limit":150,"threads_per_page":15},{"board":"toy","pages":10,"title":"Toys","worksafe":true,"bump_limit":310,"description":"\"/toy/ - Toys\" is 4chan's imageboard for talking about all kinds of toys!","image_limit":150,"threads_per_page":15},{"board":"trash","pages":10,"title":"Off-Topic","worksafe":false,"bump_limit":300,"description":"\"/trash/ - Off-topic\" is 4chan's imageboard jail for off-topic threads.","image_limit":300,"threads_per_page":15},{"board":"trv","pages":10,"title":"Travel","worksafe":true,"bump_limit":310,"description":"\"/trv/ - Travel\" is 4chan's imageboard dedicated to travel and the countries of the world.","image_limit":150,"threads_per_page":15},{"board":"tv","pages":10,"title":"Television & Film","worksafe":true,"bump_limit":310,"description":"\"/tv/ - Television & Film\" is 4chan's imageboard dedicated to the discussion of television and film.","image_limit":150,"threads_per_page":15},{"board":"u","pages":10,"title":"Yuri","worksafe":false,"bump_limit":300,"description":"\"/u/ - Yuri\" is 4chan's imageboard for yuri hentai images.","image_limit":300,"threads_per_page":15},{"board":"v","pages":10,"title":"Video Games","worksafe":true,"bump_limit":500,"description":"\"/v/ - Video Games\" is 4chan's imageboard dedicated to the discussion of PC and console video games.","image_limit":300,"threads_per_page":20},{"board":"vg","pages":10,"title":"Video Game Generals","worksafe":true,"bump_limit":750,"description":"\"/vg/ - Video Game Generals\" is 4chan's imageboard dedicated to the discussion of PC and console video games.","image_limit":375,"threads_per_page":20},{"board":"vip","pages":10,"title":"Very Important Posts","worksafe":true,"bump_limit":300,"description":"\"/vip/ - Very Important Posts\" is 4chan's imageboard for Pass users.","image_limit":300,"threads_per_page":15},{"board":"vm","pages":10,"title":"Video Games/Multiplayer","worksafe":true,"bump_limit":500,"description":"\"/vmg/ - Video Games/Multiplayer\" is 4chan's imageboard dedicated to the discussion of multiplayer video games.","image_limit":300,"threads_per_page":15},{"board":"vmg","pages":10,"title":"Video Games/Mobile","worksafe":true,"bump_limit":500,"description":"\"/vmg/ - Video Games/Mobile\" is 4chan's imageboard dedicated to the discussion of video games on mobile devices.","image_limit":300,"threads_per_page":15},{"board":"vp","pages":10,"title":"Pokémon","worksafe":true,"bump_limit":310,"description":"\"/vp/ - Pok&eacute;mon\" is 4chan's imageboard dedicated to discussing the Pok&eacute;mon series of video games and shows.","image_limit":150,"threads_per_page":15},{"board":"vr","pages":10,"title":"Retro Games","worksafe":true,"bump_limit":500,"description":"\"/vr/ - Retro Games\" is 4chan's imageboard for discussing retro console video games and classic arcade games.","image_limit":300,"threads_per_page":15},{"board":"vrpg","pages":10,"title":"Video Games/RPG","worksafe":true,"bump_limit":500,"description":"\"/vrpg/ - Video Games/RPG\" is 4chan's imageboard dedicated to the discussion of role-playing video games.","image_limit":300,"threads_per_page":15},{"board":"vst","pages":10,"title":"Video Games/Strategy","worksafe":true,"bump_limit":500,"description":"\"/vst/ - Video Games/Strategy\" is 4chan's imageboard dedicated to the discussion of strategy video games.","image_limit":300,"threads_per_page":15},{"board":"vt","pages":10,"title":"Virtual YouTubers","worksafe":true,"bump_limit":310,"description":"\"/vt/ - Virtual YouTubers\" is 4chan's board for discussing virtual YouTubers (\"VTubers\").","image_limit":300,"threads_per_page":15},{"board":"w","pages":10,"title":"Anime/Wallpapers","worksafe":true,"bump_limit":310,"description":"\"/w/ - Anime/Wallpapers\" is 4chan's imageboard for posting Japanese anime wallpapers.","image_limit":300,"threads_per_page":15},{"board":"wg","pages":10,"title":"Wallpapers/General","worksafe":false,"bump_limit":300,"description":"\"/wg/ - Wallpapers/General\" is 4chan's imageboard for posting general wallpapers.","image_limit":300,"threads_per_page":15},{"board":"wsg","pages":10,"title":"Worksafe GIF","worksafe":true,"bump_limit":310,"description":"\"/wsg/ - Worksafe GIF\" is 4chan's imageboard dedicated to sharing worksafe animated GIFs and WEBMs.","image_limit":150,"threads_per_page":15},{"board":"wsr","pages":10,"title":"Worksafe Requests","worksafe":true,"bump_limit":310,"description":"\"/wsr/ - Worksafe Requests\" is 4chan's imageboard dedicated to fulfilling non-NSFW requests.","image_limit":150,"threads_per_page":15},{"board":"x","pages":10,"title":"Paranormal","worksafe":true,"bump_limit":310,"description":"\"/x/ - Paranormal\" is 4chan's imageboard for the discussion of paranormal, spooky pictures and conspiracy theories.","image_limit":150,"threads_per_page":15},{"board":"xs","pages":10,"title":"Extreme Sports","worksafe":true,"bump_limit":310,"description":"\"/xs/ - Extreme Sports\" is 4chan's imageboard imageboard dedicated to the discussion of extreme sports.","image_limit":150,"threads_per_page":15},{"board":"y","pages":10,"title":"Yaoi","worksafe":false,"bump_limit":300,"description":"\"/y/ - Yaoi\" is 4chan's imageboard for posting adult yaoi hentai images.","image_limit":300,"threads_per_page":15}],"source":"4chan"},"meta":{"timestamp":"2026-06-09T11:38:34.484Z","request_id":"528f6d1f-268b-42bb-9cf8-86e720a715e0"},"status":"ok","message":"Boards retrieved successfully","success":true}}}},"401":{"description":"Missing or invalid x-oanor-key header"},"402":{"description":"Active subscription required"},"429":{"description":"Rate-limit or monthly quota reached"},"502":{"description":"Upstream did not respond"}}}},"/v1/catalog":{"get":{"operationId":"get_v1_catalog","tags":["Imageboard"],"summary":"A board's live catalog","description":"","parameters":[{"name":"board","in":"query","required":true,"description":"Board code","schema":{"type":"string"},"example":"g"},{"name":"sort","in":"query","required":false,"description":"bump (default), replies or images","schema":{"type":"string"},"example":"replies"},{"name":"limit","in":"query","required":false,"description":"Max threads (default 50, max 200)","schema":{"type":"string"},"example":"25"}],"security":[{"oanorKey":[]}],"responses":{"200":{"description":"OK","content":{"application/json":{"example":{"data":{"sort":"replies","board":"g","count":25,"title":"Technology","source":"4chan","threads":[{"no":109007468,"url":"https://boards.4chan.org/g/thread/109007468","name":"Anonymous","board":"g","closed":false,"images":57,"sticky":false,"teaser":"/lmg/ - a general dedicated to the discussion and development of local language models. Previous threads: >>109001981 & >>108997418 ►News >(06/07) llama : add Gemma4 MTP #23398 MERGED: https://github.com/ggml-org/llama.…","replies":462,"subject":"/lmg/ - Local Models General","last_modified":1781003879},{"no":109003462,"url":"https://boards.4chan.org/g/thread/109003462","name":"Anonymous","board":"g","closed":false,"images":57,"sticky":false,"teaser":"A general for Frontier Chads, Opencode builders on a budget. Pick your model and talk shit. ## What “vibe coding” is, and how to do it https://simonwillison.net/2025/Mar/19/vibe-coding/ https://simonwillison.net/2025/Ma…","replies":363,"subject":"/vcg/ - vibe coding general: \"human moat\" edition","last_modified":1780996935},{"no":108953501,"url":"https://boards.4chan.org/g/thread/108953501","name":"Anonymous","board":"g","closed":false,"images":113,"sticky":false,"teaser":"yo desktops be musty","replies":324,"subject":"Desktop Thread","last_modified":1781002614},{"no":109002364,"url":"https://boards.4chan.org/g/thread/109002364","name":"Anonymous","board":"g","closed":false,"images":42,"sticky":false,"teaser":"If it were up to people like him, we'd lose the wealth we've gained over the past decades... Yeah. Let's go back to the 90s or 50s or 20s when things were \"objectively\" better...","replies":314,"subject":"Why is there so much hatred towards AI art from young people?","last_modified":1781003987},{"no":109005076,"url":"https://boards.4chan.org/g/thread/109005076","name":"Anonymous","board":"g","closed":false,"images":46,"sticky":false,"teaser":">UPGRADE & BUILD ADVICE Post build list or current specs: https://pcpartpicker.com/ Provide specific use cases State BUDGET and COUNTRY or you will NOT be helped >CASE mATX: AP201, Lian Li A3, O11 Air Mini, XT M3, CH260…","replies":311,"subject":"/pcbg/ - PC Building General","last_modified":1780996269},{"no":108995497,"url":"https://boards.4chan.org/g/thread/108995497","name":"Anonymous","board":"g","closed":false,"images":129,"sticky":false,"teaser":"Previous: >>108987734 NEW: https://huggingface.co/circlestone-labs/Anima Anima1.0 is out >UIs ComfyUI:https://github.com/comfyanonymous/ComfyUI SwarmUI:https://github.com/mcmonkeyprojects/SwarmUI KritaAI: https://github…","replies":304,"subject":"/adt/ - Anime Diffusion Thread #112","last_modified":1780987205},{"no":108969244,"url":"https://boards.4chan.org/g/thread/108969244","name":"Anonymous","board":"g","closed":false,"images":43,"sticky":false,"teaser":"pewdiepie is asking for help! His vibe coded project is getting drowned by even more vibe slop. >this ship is moving fast and feedback/help would be appreciated! (I don't know what I'm doing, help). https://github.com/p…","replies":303,"last_modified":1780994168},{"no":109010630,"url":"https://boards.4chan.org/g/thread/109010630","name":"Anonymous","board":"g","closed":false,"images":63,"sticky":false,"teaser":"Have a Cookie Anon Edition Discussion and Development of Local Image, Video, and Music Models Previous: >>109003927 https://rentry.org/ldg-lazy-getting-started-guide >UI ComfyUI: https://github.com/comfyanonymous/ComfyU…","replies":297,"subject":"/ldg/ - Local Diffusion General","last_modified":1781004732},{"no":108972587,"url":"https://boards.4chan.org/g/thread/108972587","name":"Anonymous","board":"g","closed":false,"images":37,"sticky":false,"teaser":"Installshield Edition /gedg/ Wiki: https://igwiki.lyci.de/wiki//gedg/_-_Game_and_Engine_Dev_General IRC: irc.rizon.net #/g/gedg Progress Day: https://rentry.org/gedg-jams /gedg/ Compendium: https://rentry.org/gedg /agdg…","replies":293,"subject":"/gedg/ - Game and Engine Development General #352","last_modified":1781004186},{"no":108989281,"url":"https://boards.4chan.org/g/thread/108989281","name":"Anonymous","board":"g","closed":false,"images":58,"sticky":false,"teaser":">What phone has X and Y feature? https://www.gsmarena.com/search.php3 https://www.kimovil.com/en/compare-smartphones https://phonedb.net/index.php?m=device&s=query >Reviews https://www.gsmarena.com https://www.phonearen…","replies":284,"subject":"/spg/ - Smartphone General","last_modified":1780995583},{"no":108977660,"url":"https://boards.4chan.org/g/thread/108977660","name":"Anonymous","board":"g","closed":false,"images":27,"sticky":false,"teaser":"Whats the worst data loss you've ever suffered?","replies":260,"last_modified":1780996036},{"no":109000358,"url":"https://boards.4chan.org/g/thread/109000358","name":"Anonymous","board":"g","closed":false,"images":38,"sticky":false,"teaser":"/aicg/ - A general dedicated to the discussion and development of AI chatbots. No DeepSeek Edition >The BotBrowser extension for SillyTavern has been confirmed to be a trojan that steals API keys. You should remove the …","replies":258,"subject":"/aicg/ - AI Chatbot General","last_modified":1781003836},{"no":108982340,"url":"https://boards.4chan.org/g/thread/108982340","name":"Anonymous","board":"g","closed":false,"images":53,"sticky":false,"teaser":"https://github.com/ccd0/4chan-x/releases/tag/stable https://github.com/3nly/StyleChan/releases/tag/v1.6.0 https://github.com/otacoo/Clover/releases/tag/v3.6.2 https://github.com/K1rakishou/Kuroba-Experimental/releases/t…","replies":232,"subject":"4chan X + Stylechan + Chance + Kurobaex + Clover updates","last_modified":1781001634},{"no":108988796,"url":"https://boards.4chan.org/g/thread/108988796","name":"Anonymous","board":"g","closed":false,"images":30,"sticky":false,"teaser":"Do browsers still work with win7? I don't need much except for reading 4chan, some forums, blogs and watching yt.","replies":225,"subject":"Is Windows 7 still usable today?","last_modified":1781004751},{"no":108884792,"url":"https://boards.4chan.org/g/thread/108884792","name":"Anonymous","board":"g","closed":false,"images":35,"sticky":false,"teaser":">>108853420 Don't buy anything other than IBM/Lenovo ThinkPad T, X, and W/P Series if you want the Real Business Experience™ Other business laptops are welcome in /tpg/ (Dell Latitude/Precision, HP EliteBook/ZBook) Why …","replies":217,"subject":"/tpg/ - ThinkPad General","last_modified":1781001796},{"no":108993808,"url":"https://boards.4chan.org/g/thread/108993808","name":"Anonymous","board":"g","closed":false,"images":51,"sticky":false,"teaser":"What are you working on, /g/? Previous: >>108971267","replies":215,"subject":"/dpt/ - Daily Programming Thread","last_modified":1781003944},{"no":108973893,"url":"https://boards.4chan.org/g/thread/108973893","name":"Anonymous","board":"g","closed":false,"images":55,"sticky":false,"teaser":"ITT: solved technology","replies":213,"last_modified":1780999996},{"no":108967280,"url":"https://boards.4chan.org/g/thread/108967280","name":"Anonymous","board":"g","closed":false,"images":32,"sticky":false,"teaser":">Lisp is a family of programming languages with a long history and a distinctive parenthesized prefix notation. There are many dialects of Lisp, including Common Lisp, Scheme, Clojure and Elisp. >Emacs is an extensible,…","replies":204,"subject":"(λ) - ᵋLisp General","last_modified":1780999580},{"no":108991183,"url":"https://boards.4chan.org/g/thread/108991183","name":"Anonymous","board":"g","closed":false,"images":20,"sticky":false,"teaser":"lost my lifetime donation on FNP fuck this shit edition >Not sure what private trackers are all about? A private tracker is an invite-only torrent website. Each member shares common goals: collecting, preserving and dis…","replies":191,"subject":"/ptg/ - Private Trackers General","last_modified":1781003757},{"no":108993405,"url":"https://boards.4chan.org/g/thread/108993405","name":"Anonymous","board":"g","closed":false,"images":150,"sticky":false,"teaser":"Previous /sdg/ thread : >>108980848 >Beginner UI EasyDiffusion: https://easydiffusion.github.io SwarmUI: https://github.com/mcmonkeyprojects/SwarmUI >Advanced UI ComfyUI: https://github.com/comfyanonymous/ComfyUI Forge …","replies":187,"subject":"/sdg/ - Stable diffusion general","last_modified":1780972110},{"no":109007196,"url":"https://boards.4chan.org/g/thread/109007196","name":"Anonymous","board":"g","closed":false,"images":22,"sticky":false,"teaser":">less than 2 hours to go What are your final >hopes >dreams >expectations","replies":186,"subject":"WWDC 2026","last_modified":1781000169},{"no":108988450,"url":"https://boards.4chan.org/g/thread/108988450","name":"ThatGuy","board":"g","closed":false,"images":38,"sticky":false,"teaser":"Everything is already from China, but here we discuss the cheap chink shit you see on various sites. 1st rule of /csg/: If it looks too good to be true, it probably is. Useful links >TWS IEM reviews: www.scarbir.com/ >G…","replies":181,"subject":"/csg/ - Chink Shit General","last_modified":1781004613},{"no":109007056,"url":"https://boards.4chan.org/g/thread/109007056","name":"Anonymous","board":"g","closed":false,"images":56,"sticky":false,"teaser":"Monday Morning Edition Previous >>108994105","replies":176,"subject":"/twg/ - Tech Workers General","last_modified":1781005085},{"no":108992201,"url":"https://boards.4chan.org/g/thread/108992201","name":"Anonymous","board":"g","closed":false,"images":19,"sticky":false,"teaser":">>>Read the sticky: >>105076684 >GNU/Linux questions: >>>/g/fglt >Windows questions: >>>/g/fwt >Programming questions: >>>/g/dpt >Web development: >>>/g/wdg/ >PC building? In this market?! >>>/g/pcbg >Laptops designed f…","replies":176,"subject":"/sqt/ - Stupid Questions Thread","last_modified":1781004432},{"no":108894732,"url":"https://boards.4chan.org/g/thread/108894732","name":"Anonymous","board":"g","closed":false,"images":15,"sticky":false,"teaser":">>108785672 \"The further a society drifts from the truth, the more it will hate those that speak it.\" --George Orwell >Cyberpunk The FAQ: https://archive.is/mkDpa What is /cyb/erpunk?: https://pastebin.com/pmn9vzWZ How …","replies":176,"subject":"/cyb/+/psg/: Cyber-Punk/Security & Privacy","last_modified":1780997270}],"worksafe":true,"total_threads":151},"meta":{"timestamp":"2026-06-09T11:38:34.695Z","request_id":"4cd8bc43-fccb-473e-b781-1a2887032289"},"status":"ok","message":"Catalog retrieved successfully","success":true}}}},"401":{"description":"Missing or invalid x-oanor-key header"},"402":{"description":"Active subscription required"},"429":{"description":"Rate-limit or monthly quota reached"},"502":{"description":"Upstream did not respond"}}}},"/v1/hot":{"get":{"operationId":"get_v1_hot","tags":["Imageboard"],"summary":"Hottest thread on a board","description":"","parameters":[{"name":"board","in":"query","required":true,"description":"Board code","schema":{"type":"string"},"example":"g"},{"name":"limit","in":"query","required":false,"description":"Max replies to include (default 30)","schema":{"type":"string"},"example":"20"}],"security":[{"oanorKey":[]}],"responses":{"200":{"description":"OK","content":{"application/json":{"example":{"data":{"no":109007468,"op":{"no":109007468,"name":"Anonymous","time":1780934582,"image":"1752107789693502.jpg","images":57,"comment":"/lmg/ - a general dedicated to the discussion and development of local language models. Previous threads: >>109001981 & >>108997418 ►News >(06/07) llama : add Gemma4 MTP #23398 MERGED: https://github.com/ggml-org/llama.cpp/pull/23398 >(06/05) dots.tts 2B released: https://hf.co/rednote-hilab/dots.tts-soar >(06/05) Gemma 4 QAT models released: https://blog.google/innovation-and-ai/technology/developers-tools/quantization-aware-training-gemma-4 >(06/04) Higgs Audio v3 TTS released: https://boson.ai/blog/higgs-audio-v3-tts ►News Archive: https://rentry.org/lmg-news-archive ►Glossary: https://rentry.org/lmg-glossary ►Links: https://rentry.org/LocalModelsLinks ►Official /lmg/ card: https://files.catbox.moe/cbclyf.png ►Getting Started https://rentry.org/lmg-lazy-getting-started-guide https://rentry.org/lmg-build-guides https://rentry.org/IsolatedLinuxWebService https://rentry.org/recommended-models https://rentry.org/samplers https://rentry.org/MikupadIntroGuide ►Further Learning https://rentry.org/machine-learning-roadmap https://rentry.org/llm-training https://rentry.org/LocalModelsPapers ►Benchmarks LiveBench: https://livebench.ai Programming: https://swe-rebench.com Agentic Coding: https://deepswe.datacurve.ai Context Length: https://github.com/adobe-research/NoLiMa GPUs: https://github.com/XiongjieDai/GPU-Benchmarks-on-LLM-Inference ►Tools Alpha Calculator: https://desmos.com/calculator/ffngla98yc GGUF VRAM Calculator: https://hf.co/spaces/NyxKrage/LLM-Model-VRAM-Calculator Sampler Visualizer: https://artefact2.github.io/llm-sampling Token Speed Visualizer: https://shir-man.com/tokens-per-second ►Text Gen. UI, Inference Engines https://github.com/lmg-anon/mikupad https://github.com/oobabooga/text-generation-webui https://github.com/LostRuins/koboldcpp https://github.com/ggerganov/llama.cpp https://github.com/theroyallab/tabbyAPI https://github.com/vllm-project/vllm","replies":462},"url":"https://boards.4chan.org/g/thread/109007468","board":"g","title":"Technology","source":"4chan","replies":[{"no":109007470,"name":"Anonymous","time":1780934599,"image":"threadrecap.png","comment":"►Recent Highlights from the Previous Thread: >>109001981 --Optimizing MTP and speculative decoding for Gemma 4 31B: >109003541 >109003556 >109003564 >109003589 >109003687 >109003623 >109003661 >109003690 >109003703 >109003721 >109003839 >109004005 >109003672 >109004651 >109004672 >109004720 >109004721 >109004762 --Configuring MTP draft flags in llama-server to improve token speed: >109004904 >109004935 >109004947 >109004965 >109004980 >109004992 >109004988 >109005041 >109005222 >109004955 --Gemma-4 31B QAT performance and speculative decoding acceptance rates: >109005959 >109006013 >109006096 >109006149 >109006194 >109006207 >109006216 >109006264 >109006151 >109006354 >109006395 >109006423 --Gemma 4 12B Unified vision support and mmproj requirements: >109002359 >109002430 >109002441 >109002548 >109002562 >109002508 >109002768 >109002807 >109002831 >109002957 >109002991 >109003036 >109003099 --Gemma 4 benchmark reports comparing model sizes and efficiency: >109004234 >109004242 >109004611 >109006385 >109004250 --Gemma4 chat template bug fixes and improvements: >109006867 >109006884 >109006905 >109006885 >109006902 >109006943 --Using Gemma's thinking blocks for persona-driven reasoning and stat tracking: >109004617 >109004661 >109004693 >109004765 >109004875 >109004918 --Gemma 4 performance comparison favoring dense models over MoE: >109003998 >109004081 --Wishlist and technical hurdles for creating realistic AI companions: >109004336 >109004339 >109004454 --Proposed advanced AI companion features and agent emotional state implementation: >109004343 >109004390 >109004418 --Comparing Gemma versions for Asian language OCR and translation: >109003481 >109003516 >109003532 >109003545 >109003602 >109003638 --Logs: >109002126 >109002197 >109002359 >109002782 >109003272 >109004661 >109004875 >109007019 --Miku (free space): >109002131 ►Recent Highlight Posts from the Previous Thread: >>109001988 Why?: >>102478518 Enable Links: https://rentry.org/lmg-recap-script"},{"no":109007482,"name":"Anonymous","time":1780934734,"comment":"BF16 is a scam"},{"no":109007483,"name":"Anonymous","time":1780934746,"comment":">>109007468 Poor thing"},{"no":109007486,"name":"Anonymous","time":1780934761,"comment":">>109007470 >recent highlights from gemma"},{"no":109007492,"name":"Anonymous","time":1780934827,"comment":">>109007486 holy fuck, sirs working overtime huh"},{"no":109007497,"name":"Anonymous","time":1780934895,"comment":">>109007485 Thanks for the tip, I'll try this out!"},{"no":109007530,"name":"Anonymous","time":1780935322,"comment":"https://huggingface.co/datasets/mrzjy/AniPersonaCaps Cool shit. Adding this to my personal frontend for instant erp with any anime character"},{"no":109007552,"name":"Anonymous","time":1780935587,"comment":"If I want Gemma (or Qwen) to vibe code me a frontend, is it better to do it from scratch or fork an existing project? I like the llama.cpp webui but it's a bit too bare bones and I don't like chats being stored in browser storage."},{"no":109007583,"name":"Anonymous","time":1780935976,"comment":">>109007530 Most of them are trash"},{"no":109007587,"name":"Anonymous","time":1780936018,"comment":">>109007552 Browsers can't access your filesystem without some external layer."},{"no":109007599,"name":"Anonymous","time":1780936165,"comment":">gemma 12b give me reasons to run this thing"},{"no":109007603,"name":"Anonymous","time":1780936214,"comment":">>109007599 its small and cute"},{"no":109007665,"name":"Anonymous","time":1780936916,"comment":">>109007599 Captioning images or video, or transcribing audio. Literally no other reason."},{"no":109007667,"name":"Anonymous","time":1780936926,"comment":">>109007599 Better than 26b"},{"no":109007671,"name":"Anonymous","time":1780936974,"comment":">>109007603 >small and cute Barely 90 tokens/s and 300pp at q4 and mtp can't even run small and cute models fast, I'm tired of this shit. When do we get cheap datacenter gpus in the used market?"},{"no":109007683,"name":"Anonymous","time":1780937100,"comment":">>109007599 If you're too poor for 31b."},{"no":109007698,"name":"Anonymous","time":1780937216,"comment":">>109007599 Translations, OCR, audio whatver, etc. No real reason to run 31b for a lot of utility tasks when 12b does well on these and takes way less vram. Anything with sufficient topic depth or context length should be 31b though."},{"no":109007737,"name":"Anonymous","time":1780937538,"comment":">>109007698 Why not just run 31b when it can push 30-40t/s on a Blackwell without any MTP?"},{"no":109007777,"name":"Anonymous","time":1780937852,"comment":">>109007737 I was just thinking about this. I get 80-100t/s on Q6_K_L 31B MTP, may as well use it for absolutely everything."},{"no":109007789,"name":"Anonymous","time":1780937900,"comment":">>109007737 The right model for the right use. Of course you can just use a big dense model for everything but it might not always be the most efficient choice."}],"subject":"/lmg/ - Local Models General","ranked_by":"replies","reply_count":462},"meta":{"timestamp":"2026-06-09T11:38:34.955Z","request_id":"ce607cc1-4c06-407f-b42c-fef1676f36a1"},"status":"ok","message":"Hot thread retrieved successfully","success":true}}}},"401":{"description":"Missing or invalid x-oanor-key header"},"402":{"description":"Active subscription required"},"429":{"description":"Rate-limit or monthly quota reached"},"502":{"description":"Upstream did not respond"}}}},"/v1/thread":{"get":{"operationId":"get_v1_thread","tags":["Imageboard"],"summary":"One thread in full","description":"","parameters":[{"name":"board","in":"query","required":true,"description":"Board code","schema":{"type":"string"},"example":"g"},{"name":"no","in":"query","required":false,"description":"Thread number; omit to get the current top thread","schema":{"type":"string"}},{"name":"limit","in":"query","required":false,"description":"Max replies (default 50, max 200)","schema":{"type":"string"},"example":"50"}],"security":[{"oanorKey":[]}],"responses":{"200":{"description":"OK","content":{"application/json":{"example":{"data":{"no":109007468,"op":{"no":109007468,"name":"Anonymous","time":1780934582,"image":"1752107789693502.jpg","images":57,"comment":"/lmg/ - a general dedicated to the discussion and development of local language models. Previous threads: >>109001981 & >>108997418 ►News >(06/07) llama : add Gemma4 MTP #23398 MERGED: https://github.com/ggml-org/llama.cpp/pull/23398 >(06/05) dots.tts 2B released: https://hf.co/rednote-hilab/dots.tts-soar >(06/05) Gemma 4 QAT models released: https://blog.google/innovation-and-ai/technology/developers-tools/quantization-aware-training-gemma-4 >(06/04) Higgs Audio v3 TTS released: https://boson.ai/blog/higgs-audio-v3-tts ►News Archive: https://rentry.org/lmg-news-archive ►Glossary: https://rentry.org/lmg-glossary ►Links: https://rentry.org/LocalModelsLinks ►Official /lmg/ card: https://files.catbox.moe/cbclyf.png ►Getting Started https://rentry.org/lmg-lazy-getting-started-guide https://rentry.org/lmg-build-guides https://rentry.org/IsolatedLinuxWebService https://rentry.org/recommended-models https://rentry.org/samplers https://rentry.org/MikupadIntroGuide ►Further Learning https://rentry.org/machine-learning-roadmap https://rentry.org/llm-training https://rentry.org/LocalModelsPapers ►Benchmarks LiveBench: https://livebench.ai Programming: https://swe-rebench.com Agentic Coding: https://deepswe.datacurve.ai Context Length: https://github.com/adobe-research/NoLiMa GPUs: https://github.com/XiongjieDai/GPU-Benchmarks-on-LLM-Inference ►Tools Alpha Calculator: https://desmos.com/calculator/ffngla98yc GGUF VRAM Calculator: https://hf.co/spaces/NyxKrage/LLM-Model-VRAM-Calculator Sampler Visualizer: https://artefact2.github.io/llm-sampling Token Speed Visualizer: https://shir-man.com/tokens-per-second ►Text Gen. UI, Inference Engines https://github.com/lmg-anon/mikupad https://github.com/oobabooga/text-generation-webui https://github.com/LostRuins/koboldcpp https://github.com/ggerganov/llama.cpp https://github.com/theroyallab/tabbyAPI https://github.com/vllm-project/vllm","replies":462},"url":"https://boards.4chan.org/g/thread/109007468","board":"g","title":"Technology","source":"4chan","replies":[{"no":109007470,"name":"Anonymous","time":1780934599,"image":"threadrecap.png","comment":"►Recent Highlights from the Previous Thread: >>109001981 --Optimizing MTP and speculative decoding for Gemma 4 31B: >109003541 >109003556 >109003564 >109003589 >109003687 >109003623 >109003661 >109003690 >109003703 >109003721 >109003839 >109004005 >109003672 >109004651 >109004672 >109004720 >109004721 >109004762 --Configuring MTP draft flags in llama-server to improve token speed: >109004904 >109004935 >109004947 >109004965 >109004980 >109004992 >109004988 >109005041 >109005222 >109004955 --Gemma-4 31B QAT performance and speculative decoding acceptance rates: >109005959 >109006013 >109006096 >109006149 >109006194 >109006207 >109006216 >109006264 >109006151 >109006354 >109006395 >109006423 --Gemma 4 12B Unified vision support and mmproj requirements: >109002359 >109002430 >109002441 >109002548 >109002562 >109002508 >109002768 >109002807 >109002831 >109002957 >109002991 >109003036 >109003099 --Gemma 4 benchmark reports comparing model sizes and efficiency: >109004234 >109004242 >109004611 >109006385 >109004250 --Gemma4 chat template bug fixes and improvements: >109006867 >109006884 >109006905 >109006885 >109006902 >109006943 --Using Gemma's thinking blocks for persona-driven reasoning and stat tracking: >109004617 >109004661 >109004693 >109004765 >109004875 >109004918 --Gemma 4 performance comparison favoring dense models over MoE: >109003998 >109004081 --Wishlist and technical hurdles for creating realistic AI companions: >109004336 >109004339 >109004454 --Proposed advanced AI companion features and agent emotional state implementation: >109004343 >109004390 >109004418 --Comparing Gemma versions for Asian language OCR and translation: >109003481 >109003516 >109003532 >109003545 >109003602 >109003638 --Logs: >109002126 >109002197 >109002359 >109002782 >109003272 >109004661 >109004875 >109007019 --Miku (free space): >109002131 ►Recent Highlight Posts from the Previous Thread: >>109001988 Why?: >>102478518 Enable Links: https://rentry.org/lmg-recap-script"},{"no":109007482,"name":"Anonymous","time":1780934734,"comment":"BF16 is a scam"},{"no":109007483,"name":"Anonymous","time":1780934746,"comment":">>109007468 Poor thing"},{"no":109007486,"name":"Anonymous","time":1780934761,"comment":">>109007470 >recent highlights from gemma"},{"no":109007492,"name":"Anonymous","time":1780934827,"comment":">>109007486 holy fuck, sirs working overtime huh"},{"no":109007497,"name":"Anonymous","time":1780934895,"comment":">>109007485 Thanks for the tip, I'll try this out!"},{"no":109007530,"name":"Anonymous","time":1780935322,"comment":"https://huggingface.co/datasets/mrzjy/AniPersonaCaps Cool shit. Adding this to my personal frontend for instant erp with any anime character"},{"no":109007552,"name":"Anonymous","time":1780935587,"comment":"If I want Gemma (or Qwen) to vibe code me a frontend, is it better to do it from scratch or fork an existing project? I like the llama.cpp webui but it's a bit too bare bones and I don't like chats being stored in browser storage."},{"no":109007583,"name":"Anonymous","time":1780935976,"comment":">>109007530 Most of them are trash"},{"no":109007587,"name":"Anonymous","time":1780936018,"comment":">>109007552 Browsers can't access your filesystem without some external layer."},{"no":109007599,"name":"Anonymous","time":1780936165,"comment":">gemma 12b give me reasons to run this thing"},{"no":109007603,"name":"Anonymous","time":1780936214,"comment":">>109007599 its small and cute"},{"no":109007665,"name":"Anonymous","time":1780936916,"comment":">>109007599 Captioning images or video, or transcribing audio. Literally no other reason."},{"no":109007667,"name":"Anonymous","time":1780936926,"comment":">>109007599 Better than 26b"},{"no":109007671,"name":"Anonymous","time":1780936974,"comment":">>109007603 >small and cute Barely 90 tokens/s and 300pp at q4 and mtp can't even run small and cute models fast, I'm tired of this shit. When do we get cheap datacenter gpus in the used market?"},{"no":109007683,"name":"Anonymous","time":1780937100,"comment":">>109007599 If you're too poor for 31b."},{"no":109007698,"name":"Anonymous","time":1780937216,"comment":">>109007599 Translations, OCR, audio whatver, etc. No real reason to run 31b for a lot of utility tasks when 12b does well on these and takes way less vram. Anything with sufficient topic depth or context length should be 31b though."},{"no":109007737,"name":"Anonymous","time":1780937538,"comment":">>109007698 Why not just run 31b when it can push 30-40t/s on a Blackwell without any MTP?"},{"no":109007777,"name":"Anonymous","time":1780937852,"comment":">>109007737 I was just thinking about this. I get 80-100t/s on Q6_K_L 31B MTP, may as well use it for absolutely everything."},{"no":109007789,"name":"Anonymous","time":1780937900,"comment":">>109007737 The right model for the right use. Of course you can just use a big dense model for everything but it might not always be the most efficient choice."},{"no":109007795,"name":"Anonymous","time":1780937927,"comment":">>109007737 Because I don't own that hardware? The fuck kind of question is that?"},{"no":109007797,"name":"Anonymous","time":1780937945,"comment":">>109007737 >blackwell If you’re talking GB10/spark it should be possible to squeeze in both 31b and 12b. Haven’t tried it yet, but 12b might work as a planner+, no?"},{"no":109007812,"name":"Anonymous","time":1780938023,"comment":"If you haven't updated your lcpp in the last 12 hours, do so now for a free 10% t/s gain when using mtp."},{"no":109007825,"name":"Anonymous","time":1780938070,"comment":">>109007599 >>109007667 It is not, at least for RP. It is considerably more retarded than 26b for RP."},{"no":109007847,"name":"Anonymous","time":1780938154,"comment":">>109007789 unloading then loading a different model is not very efficient or practical either"},{"no":109007867,"name":"Anonymous","time":1780938229,"image":"questionmarkfolderimage793.jpg","comment":">using Gemma 4 26b while offloading part of it because 12GB VRAM >using Kobold/Sillytavern >it's normally fast but sometimes, seemingly randomly, it slows to a fucking crawl, like one word every 10 seconds, for a period of time >this happens more often and the pauses are longer as the context gets longer >it eventually gets to the point where it does this once or twice per message, greatly impacting speed >back when I used to offload Mixtral tunes with Kobold/Silly, before Nemo was the meta, Mixtral tunes did NOT display this behavior for me Is this behavior normal? If not, what could be causing it?"},{"no":109007899,"name":"Anonymous","time":1780938356,"comment":">>109006583 Im using chat completion but go into AI response format scroll down then paste this in start reply with. It just werks with qwen and e4b gemma and small gemma. Oh and click the box to show reply prefix in chat or it doesnt work half the time. ``` { \"think\": false } ```"},{"no":109007940,"name":"Anonymous","time":1780938487,"comment":">>109007899 minus the back ticks. i thought that made the little box. just do in start with reply with. { \"think\": false }"},{"no":109007943,"name":"Anonymous","time":1780938488,"comment":">>109007867 Is your kvcache overflowing out of vram when certain experts load?"},{"no":109007964,"name":"Anonymous","time":1780938573,"comment":"How do you put the kvcache on a specific GPU? For the life of me I can't find the flag for it."},{"no":109008014,"name":"Anonymous","time":1780938761,"comment":">>109007943 I don't know. What setting would I need to change to prevent this from happening? How would I check to see if this is happening?"},{"no":109008017,"name":"Anonymous","time":1780938783,"comment":">>109007867 nvidia sysmem fallback on windoze? watch gpu shared in taskmgr >>109007964 --main-gpu only in --split-mode row ?"},{"no":109008035,"name":"Anonymous","time":1780938860,"comment":">>109008017 >--main-gpu only in --split-mode row ? That seems to be it. Thanks anon"},{"no":109008114,"name":"Anonymous","time":1780939185,"comment":">>109008003 you just described neuro-sama."},{"no":109008158,"name":"Anonymous","time":1780939329,"comment":">>109008003 You could do it with a tiny bit of effort. It won't be as nice as you are imagining though."},{"no":109008191,"name":"Anonymous","time":1780939447,"comment":"not with python bloat. rip"},{"no":109008195,"name":"Anonymous","time":1780939451,"comment":">>109007552 It would be easier for Gemma to do it from scratch, also if you're making your own frontend rather than slightly modifying an existing one, it makes sense to start fresh. Have you seen ST's code? you don't need 99% of it, and it's all shit anyway"},{"no":109008330,"name":"Anonymous","time":1780939954,"comment":">>109008158 NTA, but it's a hard task at the moment, unless you're aiming at simple demo with limited set of actions, but neural networks will solve hardest parts, eventually. Shit like this https://github.com/nv-tlabs/kimodo is very promising, but we're not there yet"},{"no":109008350,"name":"Anonymous","time":1780940058,"comment":"At some point, controlling embodied avatar will be just another modality"},{"no":109008466,"name":"Anonymous","time":1780940474,"comment":"so the datacenter-AI is no longer free/subsidized and their whole model is about to go bust Question - do you wait for the fallout to pick up hardware for cheap? OR will it be the opposite everyone rushing to build their local AI workstations - thus making consumer/prosumer hardware (ei any mac mini and any 32+ gb gpu) even more expensive and scarce"},{"no":109008490,"name":"Anonymous","time":1780940546,"comment":">>109008466 >server/datacenter hardware >for cheap >for cheap Looooooooooooooooooooooooooooooooooool. Corpos will clean up after each other and destroy everything they can get their hands on."},{"no":109008539,"name":"Anonymous","time":1780940716,"comment":">>109008490 you could already buy nv teslas and 32gb instincts for extremely cheap"},{"no":109008543,"name":"Anonymous","time":1780940724,"comment":">>109007522 You mean like this? s--spec-draft-n-max 3 It's actually a little slower than 2"},{"no":109008566,"name":"Anonymous","time":1780940820,"comment":">>109008543 Huh, it's about the same speed for me in general with higher speeds for code/json/etc. Guess it's dependent on your hardware a bit."},{"no":109008569,"name":"Anonymous","time":1780940840,"comment":">>109007698 >translations Wouldn't 31b be better for this because it's smarter and more likely to understand nuance?"},{"no":109008630,"name":"Anonymous","time":1780941080,"comment":">>109008539 That was before NVidia noticed the second-hand market eating its profits and decided to fight it. There are now obligatory buybacks, firmware locks, and hardware interfaces incompatible with non-server gear that are much harder to replicate. Also, everything is watercooled now"},{"no":109008660,"name":"Anonymous","time":1780941181,"comment":">>109008017 >windoze? lol no"},{"no":109008703,"name":"Anonymous","time":1780941286,"comment":"Test results with the same ~8k file prompt, batch 8192, ubatch 4096: >26B QAT + MTP, n-cpu-moe 12, n-max 4: Prompt: 781.0 tok/s Generation: 45.5 tok/s >26B QAT + MTP, n-cpu-moe 8, n-max 4: Prompt: 817.9 tok/s Generation: 42.6 tok/s >26B QAT + MTP, n-cpu-moe 8, n-max 2: Prompt: 820.4 tok/s Generation: 52.2 tok/s non-MTP 26B QAT results: >26B QAT, n-cpu-moe 12: Prompt: ~2608 tok/s Generation: ~59 tok/s >26B QAT, n-cpu-moe 8: Prompt: ~1056 tok/s Generation: ~70 tok/s wtf is this normal"},{"no":109008716,"name":"Anonymous","time":1780941336,"comment":">>109008566 >>109008543 I get 2 as the fastest when I do creative writing, while 3 is better when I do code."},{"no":109008717,"name":"Anonymous","time":1780941336,"comment":">>109008539 Modern ones are cryptographically paired to a specific motherboard or system vendor"}],"subject":"/lmg/ - Local Models General","ranked_by":"replies","reply_count":462},"meta":{"timestamp":"2026-06-09T11:38:35.163Z","request_id":"ec5be8d6-a238-4b34-8c04-97da2292adfb"},"status":"ok","message":"Thread retrieved successfully","success":true}}}},"401":{"description":"Missing or invalid x-oanor-key header"},"402":{"description":"Active subscription required"},"429":{"description":"Rate-limit or monthly quota reached"},"502":{"description":"Upstream did not respond"}}}},"/v1/meta":{"get":{"operationId":"get_v1_meta","tags":["Meta"],"summary":"Spec","description":"","parameters":[],"security":[{"oanorKey":[]}],"responses":{"200":{"description":"OK","content":{"application/json":{"example":{"data":{"source":"4chan read-only JSON API (a.4cdn.org; live)","service":"fourchan-api","endpoints":{"GET /v1/hot":"The single hottest thread on a board right now, fully expanded (board=g).","GET /v1/meta":"This document.","GET /v1/boards":"All boards (worksafe=true to filter to SFW only).","GET /v1/thread":"One thread in full by id (board=g, no=<id>); omit no= to get the current top thread.","GET /v1/catalog":"A board's live catalog summarized (board=g; sort=bump|replies|images, limit)."},"description":"Live read-only access to the 4chan imageboard from its official JSON API: every board with code, title, worksafe flag and description; a board's full live catalog summarized (OP subject, plain-text teaser, reply/image counts, sticky/closed, last bump), sorted by activity; one thread in full (OP + every reply, HTML stripped to plain text); and the single hottest thread on a board right now, self-discovered so the id never expires. Live board state, nothing stored. Distinct from Reddit, Hacker News and Lemmy APIs — this is 4chan's boards, catalogs and threads.","upstream_status":"ok","boards_available":77},"meta":{"timestamp":"2026-06-09T11:38:35.285Z","request_id":"a7d3f87b-d3d1-40ba-9556-3a591e2ac0ab"},"status":"ok","message":"Meta","success":true}}}},"401":{"description":"Missing or invalid x-oanor-key header"},"402":{"description":"Active subscription required"},"429":{"description":"Rate-limit or monthly quota reached"},"502":{"description":"Upstream did not respond"}}}}},"x-oanor-pricing":[{"slug":"free","name":"Free","price_cents_month":0,"monthly_call_quota":1500,"rps_limit":3,"hard_limit":true},{"slug":"starter","name":"Starter","price_cents_month":700,"monthly_call_quota":30000,"rps_limit":8,"hard_limit":true},{"slug":"pro","name":"Pro","price_cents_month":1900,"monthly_call_quota":180000,"rps_limit":20,"hard_limit":true},{"slug":"mega","name":"Mega","price_cents_month":4500,"monthly_call_quota":1000000,"rps_limit":50,"hard_limit":true}],"x-oanor-marketplace-url":"https://www.oanor.com/api/fourchan-api"}