Tools

MOSS-TTS-Local Transformer v1.5 is now released on mlx-audio and open sourced.

Ivan Fioravanti demonstrates a local Apple Silicon workflow that pairs mlx-audio, the MOSS-TTS-Local-Transformer-v1.5 text-to-speech model, and Kling Omni to generate lip-synced videos.

Darkbloom is a Mac-focused distributed private inference network with a distinctive architecture that aims to enable privacy-preserving model inference across machines.

VulcanBench v0.1 is a free, open-source LLM benchmarking toolkit (announced 2026-06-20) that ships with 52 pre-loaded tests.

The post by @echelonylol loudly expresses desire for the Loop Candleholder by NÆR.

A reply from @SuspendedDr expresses strong desire for Spacesthetic (@interiorsuckerr)'s novelty product, described as "a human spine stackable glasses," which is shown in an attached video.

El post de vicentebrisa_ expresa entusiasmo por “estas pavadas”, es decir pequeños detalles decorativos, acompañando una imagen compartida por Spacesthetic (@interiorsuckerr) del set de pufs “Toasted” de Living in Design.

On 2026-06-18, @DotCunt reshared a design titled “Spacesthetic” credited to LCMdesign and originally posted by @interiorsuckerr on X, accompanied only by the short French exclamation “J'en veux 4 !” This conveys clear enthusiasm and a desire to obtain four of the Spacesthetic items, indicating positive reception and strong purchase or collection intent.

Iain Dunning announced a 2026 revision release on 2026-06-20 that includes a 'very limited edition jibbitz.' He brands the drop as 'dripped out technology brothers', UWS edition, signaling a themed, small-batch accessory (jibbitz) tied to the UWS variant of his project or collection.

UpSkill is an open-source project announced by @huang_chao4969 on 2026-06-20, with the source code available at github.com/HKUDS/UpSkill.

On 2026-06-20 @heyshrutimishra (reshared by Nous Research) tweeted “Hermes Agent is so good right now…,” signaling user enthusiasm about Hermes Agent’s performance.

So Good that she'll NEVER Think of Another Guy again!

The post by @GoodAssSub announces the release of a project titled "BULLY DELUXE," published on Twitter/X on June 20, 2026.

BULLY DELUXE is a new release from artist @GoodAssSub, announced on X/Twitter on 2026-06-20.

Adey (Twitter handle @ItsAdey) announced on 2026-06-20 that their vinyl is available and invited followers to "catch me on 'Mission Control'," using piano and writing emojis to emphasize musical and songwriting roles and a goat emoji suggesting confidence or acclaim.

KING WALLPAPER is a wallpaper created by community member Savi and shared by the account @GoodAssSub.

Garry Tan promoted paxel.ycombinator.com on 2026-06-20 with the one-line prompt “Try paxel.ycombinator.com and find out.” In response, Hubert (@illiqid) asked whether ‘‘token maxing my codex’’ would earn a YC interview, tagging @garrytan and linking the exchange on a nitter mirror.

The tweet by @fromzerotomill (published 2026-06-20) is a brief promotional call-to-action directing readers to FanPro's exploration page at fanprostudio.ai/explore.

A Florida developer is marketing luxury “man caves” starting at about $500K — essentially oversized, outfitted garages featuring storage, golf simulators, bars, lounges, TVs, offices, gyms and room for expensive hobbies.

Author issues a $250 challenge tied to a 5-star App Store review for anyone who posts a Psycho app score higher than theirs.

A June 20, 2026 post by @fastrlife on X links directly to the Canadian Apple App Store page for an app titled beginning “Psycho…”.

The post by @fastrlife (2026-06-20) is a brief promotional call-to-action asserting that "Elite aesthetics are the key" and directing readers to begin using psychoaesthetics.app for free.

The post by @EvaLovesDesign on Twitter/X (published 2026-06-21) highlights an Etsy product titled "Tulip Crochet Bracelet." The account, which centers on design and handmade goods, appears to be promoting or showcasing this specific crochet bracelet listing on Etsy.

Velvet Drechsel Dachshund Embroidered is a 2026-06-21 Twitter/X post by @EvaLovesDesign that appears to showcase or announce an embroidered dachshund design tied to Velvet Drechsel.

AhaCreator claims a database of more than 5 million creators and 100,000+ registered users.

The 3.0 update, announced by @dr_cintas on 2026-06-20, introduces three operational upgrades: multimodal risk screening to detect brand-safety problems before campaigns launch, an automated content-review workflow to eliminate manual approval of each post, and built-in creator payments covering more than 140 countries to simplify payouts.

Author @dr_cintas promoted early access to AhaCreator on 2026-06-20, sharing the sign-up link ahacreator.com/?utm_source=a… and explicitly thanking the team for the partnership.

AhaCreator 3.0 is presented as an end-to-end AI agent for influencer marketing that automates creator discovery, vetting, campaign management, content review, and international payments.

The post highlights @dieworkwear's recent flag-themed merch as a rare example of quality in a trend the author labels 'slap a flag on it,' dated 2026-06-19.

The post identifies the knit as a Lemmermayer cardigan available from retailer Maus & Hoffman and gives fit guidance: opt for a slightly loose size so the sweater forms a gentle 'pooch' at the waist when fastened, and avoid making the garment suction-fit for a better look.

The post by @dieworkwear (2026-06-20) highlights the online RRL vintage shop as a source of desirable vintage pieces, warns that prices are high due to labor-intensive procurement and presentation, and advises using the listings’ information and photos as research tools to inform personal vintage-hunting and sourcing.

The post by @DeanTTraining promotes pairing Bulgarian split squats with the Rogers Squat Pro, framing the combo as an ideal setup for the exercise and asserting it is “the best way to do ’em.” The message is backed by a demonstration video and a linked clip from Wong (@WongEXE___) showing the Bulgarian split squat technique.

Understand Anything is an open-source plugin for agent platforms like Claude Code, Codex and OpenCode that analyzes repositories and builds an interactive knowledge base that explains the project rather than merely listing files.

A post by @daniel_mac8 on 2026-06-20 shares a GitHub repository hosted under the Egonex-AI organization (github.com/Egonex-AI/Underst…).

Charles (@charles_irl) announces a co-release with Z Lab of six state-of-the-art DFlash speculators for Alibaba Qwen 3.x, touting a measured throughput exceeding 1,000 output transactions per second for Qwen 3.5 122B-A10B on a B200.

Charles (@charles_irl) announced on 2026-06-20 that, together with collaborators Zhijian Liu (@zhijianliu_) and Jian Chen (@jianchen1799), they published models to Hugging Face.

Luke Alonso posted an NVFP4 dump of GLM 5.2 sized at 467 GB.

User @Boenau shared a bare Wiley shop link (wiley.com/en-us/shop/general…) on Twitter/X on 2026-06-20.

GLM 5.2, according to @beaverd on 2026-06-20, runs locally at no cost and reliably follows complex, explicit multi-step instructions that the author says Opus could not execute.

CodingBrice reports receiving a Reddit audit from Anthony Riera's RankHog on 2026-06-20 and is impressed: the report identifies where to begin and which angles to use for posts.

@AlpacaAurelius tweeted on 2026-06-20 that customers are seeing "insane" hair growth after using their collagen.

Zeely AI announces that its avatars have been updated to display more natural emotions and on-camera reactions to avoid 'robotic' faces.

adityavg13 shared a World Cup–themed interactive submission hosted at their GitHub Pages URL on 2026-06-20 and invited people to play it.

Dreamina Seedance 2.0 Mini is available in CapCut, emphasizing speed and cost efficiency: roughly double the generation speed and about 30% lower costs versus Seedance 2.0 Fast.

Exo supports MLX as of 2026-06-20, providing a curated collection of tools, models, and configs for users to choose from.

0xSero emphasizes reducing cost barriers for capable models: DS4-Flash can run on $2.5–4.5k setups versus prior $9k–$20k requirements.

GLM 5.2 NVFP4 REAP 469B is running on a 3× DGX Spark cluster, delivering a 256,000-token context window with about 4.4 tokens/second throughput.

GLM-5.2-REAP is praised for proactively checking its work and catching errors before delivering results, a behavior the author finds highly valuable.

Author highlights that Z.

@LKrauss1 describes a hands-off trip from South San Francisco through Oregon on June 20, 2026, crediting Tesla Full Self-Driving for enabling the journey.

RobertJBye highlights Claude Code’s new 'Artifacts' (announced 2026-06-18) as transformative for product managers: it builds interactive pages from a session—dashboards, wireframe demos, research summaries—that link to live data and are shared via private links, currently in beta for Team and Enterprise plans.

A Swiss watch escapement simulation by @Legendaryy (thread 9/, posted 2026-06-11) models gears, a hairspring, and real-time hands; it uses a vision-based self-correction loop to fix physics errors during the simulation.

Kickbacks, introduced by Andrew McCalip on 2026-06-11, monetizes the Claude code-generation spinner by inserting auctioned ads during wait times.

@kyleanthony used a brief Twitter/X post on 2026-06-19 to drive traffic to their studio portfolio at brasshands.com.

The mapgl 0.5 release adds substantial UI controls, notably a standalone slider that enables interactive client-side filtering (no Shiny required).

Kyle E.

The post by @brainlety (2026-06-19) invites anyone thinking of joining @TheKookLetter to contact them for a discussion about the newsletter’s benefits.

Justin Wolfers posted on 2026-06-19 promoting a Father's Day deal: annual Platypus gift subscriptions are 25% off.

LiquidAI released two 350M-parameter multilingual retrieval models on 2026-06-18: LFM2.5-Embedding-350M (a bi-encoder) and LFM2.5-ColBERT-350M (a multi-vector ColBERT with late interaction).

Graham Neubig reported on 2026-06-19 that GLM-5.2 is a strong open model—potentially the first viable replacement for closed models for most tasks aside from vision.

LiquidAI shows that overloading a model context with many tools increases latency and selection errors.

Coffee Collective in Copenhagen received a strong endorsement from @jasonc_nc (post dated 2026-06-19): despite losing the original recommending comment, they visited and found the coffee "absolutely excellent," explicitly comparing its quality to Nomad Coffee in Barcelona and calling them equal.

Liquid AI released two 350M-parameter retrieval models, LFM2.5-Embedding-350M and LFM2.5-ColBERT-350M, claiming end-to-end retrieval latency down to 1.5 ms and 'best-in-class' multilingual/cross-lingual performance across 11 languages.

Poolside launched Laguna M.1 on 2026-06-18, publishing both base and post-trained checkpoints on Hugging Face under Apache 2.0 and advertising a very large 256K context length; Prince Canuma (@Prince_Canuma) congratulated them and announced day‑0 support on MLX, providing the install hint `uv pip install -U mlx-vlm` to get started.

MLX-LoRA-Studio is being released in its first 'oben' version by @ActuallyIsaak (GitHub: Goekdeniz-Guelmez).

The "Toasted" pouf set is a product release from Living in Design.

User @interiorsuckerr shared a brief social post on 2026-06-19 highlighting a miniature bookmark made by creator Minyatür Dünyam.

The X post by @9rizumu dated 2026-06-18 provides a first look at a project called 'Reflections' Trainer, signaling an initial public preview.

The post from @homegymcoop (2026-06-19) highlights an adjustable kettlebell that reaches 102.5 lb and equates to $3.40 per pound at that top weight, with the recommendation that mounting your own plates makes it much cheaper.

The Revolt Adjustable KB is recommended by @homegymcoop in a short X/Twitter post dated 2026-06-19; the author presents the name as a purchase tip for fitness friends and stresses they’re not trying to spam links, simply to be helpful.

The post by @homegymcoop (2026-06-19) urges people building budget home gyms to skip foam tiles and buy Costco’s rubber mats, which the author says cost roughly $18 and are cheaper than a McDonald’s meal.

Developer @hasantoxr posted on 2026-06-19 asking followers to open and Star the GitHub repository EverMind-AI/EverO to support a rebuild, and to file feedback or issues on GitHub after starring; the message is shared with the hashtag #EverOS to rally community contributions.

AutoGTM, recommended by @hasantoxr, promises an automated GTM workflow: paste your product URL, let AI research your market in ~45 seconds, surface target companies, then generate and send personalized email outreach continuously.

@hamptonism describes the experience of an agent that pre-checks funding rates, whale movements, centralized exchange depth, and on-chain flows to give trading signals overnight.

GlobalGPT is presented as an all-in-one AI workspace (glbgpt.com) that aggregates over 100 models for $10/month.

Author @ginacostag_ announced on 2026-06-19 that GlobalGPT has made two large models—Gemini 3.5 Flash and Claude 4.5—public and free.

Claude Code's new Artifacts feature lets users deploy HTML sites directly from a session and share them with teammates via private links.

A Twitter/X post by @financedystop (2026-06-19) advertises a 2015 GMC 2500 Crew Cab with 839,930 miles for $12,500.

Claude Code introduced "Artifacts," which turn session outputs into shareable interactive pages (e.g., PR walkthroughs or living project dashboards) accessible by a private link.

The Twitter/X post by @EvaLovesDesign on 2026-06-19 is a minimal, two-word mention: "smathers and branson." The entry likely names the needlepoint accessories brand Smathers & Branson (belts and small leather goods) but provides no context, opinion, or link to indicate whether this is a recommendation, inquiry, or casual remark.

EvaLovesDesign (tweet on 2026-06-19) urges followers to purchase framed American flags whenever they find them, reiterating a previously stated tip.

The post announces Lightricks' open-source LTX-2 project with a GitHub repo (github.com/Lightricks/LTX-2/), accompanying documentation (docs.ltx.video/open-source-m…), and a Hugging Face collection (huggingface.co/collections/L…).

LTX Trainer enables non-experts to train custom video models by speaking plain-language goals to Claude; an agentic skill then generates the training config and runs the training pipeline automatically.

Derrick C.

The post by @DeanTTraining praises a Ninja Creami Deluxe–made treat as tasting and looking like a Dairy Queen item while offering substantially better nutrition: about one-third the Calories and five times the protein of the Dairy Queen option.

Trainer @DeanTTraining effusively praises a new Prime leg-extension machine posted 2026-06-19, calling it the greatest ever and lavishing hyperbolic praise on Prime’s design team.

Hermes Agent received a brief but emphatic endorsement from @DavidOndrej1 on X/Twitter on 2026-06-19, claiming it is "so good right now." The post signals high user enthusiasm or perceived improvement, though the author offered no supporting metrics, use cases, or comparisons to substantiate the claim.

RubenDinis12 reported that Polygraph.so gained the ability to grade skills installed in Cursor, posting the announcement on 2026-06-19.

Author @DataChaz, writing on 2026-06-19, says that using Microsoft Teams at major corporations a few years ago left them sympathetic to current users' struggles (accompanied by a laughing/sweat emoji).

The tweet by @DataChaz (published 2026-06-19) links to the GitHub repository github.com/chopratejas/headr, gives a public shoutout to @chopra_tejas for building the project for the community, and encourages followers to show support by starring the repository.

@DataChaz promotes HappySeeds (post dated 2026-06-19), sharing a referral link (link.happyseeds.ai/7xjq8e) and a promo code CHARLY2026 that grants 200 bonus credits.

Author @DataChaz posted the second item in a thread on 2026-06-19 titled “2/ Football demo,” directing readers to a demo application hosted at app-ce3abc4512.happyseeds.sp via a short link.

Author @DataChaz (4/ in a thread, published 2026-06-19) is soliciting questions and project showcases from followers and directs them to join a community Discord using the invite link discord.gg/hasxqvQyP for discussion and sharing.

Headroom is a free, open-source local proxy (released by a Netflix engineer) that transparently compresses payloads sent to models such as Cursor and Claude, claiming up to 95% token reduction without changing application code.

Jeremy Howard, quoted by @daniel_mac8 on 2026-06-19, praises Zai_org’s GLM 5.2 as matching or exceeding Opus 4.8 and GPT 5.5 in quality.

Claude Code Artifacts deliver instant, team-wide updates: changes made in one user’s session propagate immediately to the shared artifact for all teammates.

Eve is a framework for building Slack agents that handles app provisioning, OAuth, and event handling with minimal TypeScript.

The post describes a simple consulting model for small business owners: a done-with-you subscription charging $1,500–$2,000/month for two 45-minute strategy calls.

GLM 5.2 is drawing strong acclaim: AndrewCurran notes a flurry of positive reactions over 48 hours (June 2026), while Jeremy Howard praises Zai_org's release as matching or exceeding Opus 4.8 and GPT 5.5.

The author announced an open-source 9B document-extraction model that achieves 90.2% on their benchmark—close to Gemini 3.5 Flash’s 91.3%—and substantially ahead of NuExtract3 (81.5%).