# forck.live — 22–28 April 2024 > AI lab announcements, news reports and podcasts. forck.live follows what companies announce about their own products. Press may corroborate an official story; it never leads and never appears as an item of its own. A missing date stays missing. A title is the publisher's words — where a company does not publish in English, an English rendering is listed separately as `Title in English` and named as this site's translation, never as theirs. This file is one issue of the weekly archive: every confirmed first-party announcement of ISO week 2024-W17, 22–28 April 2024, newest first. The window never moves, so this URL always names the same seven days — but a post published inside them and detected later is added when it arrives, so a closed week can still gain entries. Every entry is one first-party announcement, and carries the same eight fields in this order before any others: Company · Announced · Category · Coverage · Announcement · Group · Source · Record `Source` is the company's own page the announcement was traced to and `Record` is this site's page for it. Both are safe to cite. `Coverage` is a number of outlets, or `not counted` — which means this record reached no count of its own, never that nobody wrote about it. `Category` is this tracker's own analysis, and is authoritative wherever it is stated. `not stated` means either that nothing has analysed the post yet or that the analysis placed it outside this taxonomy; neither is a verdict about the announcement. `Announcement` is whether the post is news of something the company did, rather than a how-to, a customer story, a changelog listing or a staffing note. `Group` is which of this record's groups the entry falls in, carried on the entry itself so it holds where the groups are headings and where they are not. One of: `models`, `covered`, `announcements`, `routine`. Anything after `Record` is held detail that only some announcements carry. - This week's page: https://forck.live/week/2024-W17 - The current issue, and the catalogue of the other weeks: https://forck.live/briefing.md - Catalogue: https://forck.live/llms.txt ## Announcements ### Vercel — How Vercel helped Tonies expand into new markets and improve conversion rates - Company: Vercel (vercel.com) - Announced: 2024-04-26T13:00:00+00:00 - Category: not stated - Coverage: not counted - Announcement: no - Group: routine - Source: https://vercel.com/blog/how-vercel-helped-tonies-expand-into-new-markets - Record: https://forck.live/items/20988-how-vercel-helped-tonies-expand-into-new-markets-and-improve-conversion-rates - Subject: v0 / AI SDK / AI Gateway Tonies, creator of a children's audio system, migrated to a frontend platform powered by Vercel and Contentful, scaling from two markets (Germany and the UK) to four markets and improving conversion rates. The team moved from infrequent deployments to multiple deployments per week and used Vercel's Preview Deployments and CDN caching to reduce latency and improve scalability during peaks like Black Friday. ### Supabase — Postgres Bloat Minimization - Company: Supabase (supabase.com) - Announced: 2024-04-26 - Category: not stated - Coverage: not counted - Announcement: no - Group: routine - Source: https://supabase.com/blog/postgres-bloat - Record: https://forck.live/items/21428-postgres-bloat-minimization - Subject: Supabase Supabase published a guide on minimizing Postgres table bloat, explaining how heap storage and vacuuming work, and how to diagnose and address autovacuum failures. The post covers tuning autovacuum parameters, using VACUUM FULL or pg_repack, and handling long-running transactions that prevent cleanup. ### Alibaba — Qwen1.5-110B: The First 100B+ Model of the Qwen1.5 Series - Company: Alibaba (alibaba.com) - Announced: 2024-04-25T05:33:00+00:00 - Category: new-model - Coverage: not counted - Announcement: yes - Group: models - Source: https://qwenlm.github.io/blog/qwen1.5-110b/ - Record: https://forck.live/items/2315-qwen1-5-110b-the-first-100b-model-of-the-qwen1-5-series - Subject: Qwen Alibaba releases Qwen1.5-110B, the first 100B+ model in the Qwen1.5 series, achieving performance comparable to Meta-Llama3-70B on base model evaluation and outstanding results on chat benchmarks. ### Supabase — Exploring Support Tooling at Supabase: A Dive into SLA Buddy - Company: Supabase (supabase.com) - Announced: 2024-04-25 - Category: not stated - Coverage: not counted - Announcement: no - Group: routine - Source: https://supabase.com/blog/exploring-support-tooling - Record: https://forck.live/items/21429-exploring-support-tooling-at-supabase-a-dive-into-sla-buddy - Subject: Supabase Supabase developed SLA Buddy, an internal support tool built on its own Edge Functions and PostgreSQL, to enforce Service Level Agreements for support tickets. The tool monitors Slack channels, checks engineer availability via Google Calendar, and escalates overdue tickets through four escalation steps. ### OpenAI — GPT-4 API general availability and deprecation of older models in the Completions API - Company: OpenAI (openai.com) - Announced: 2024-04-24 - Category: api-release - Coverage: not counted - Announcement: yes - Group: announcements - Source: https://openai.com/index/gpt-4-api-general-availability - Record: https://forck.live/items/781-gpt-4-api-general-availability-and-deprecation-of-older-models-in-the - Subject: GPT / ChatGPT / API OpenAI announced general availability of the GPT-3.5 Turbo, DALL·E, and Whisper APIs, along with a deprecation plan for older models in the Completions API that will retire in early 2024. ### OpenAI — Accelerating the development of life-saving treatments - Company: OpenAI (openai.com) - Announced: 2024-04-24 - Category: not stated - Coverage: not counted - Announcement: no - Group: routine - Source: https://openai.com/index/moderna - Record: https://forck.live/items/780-accelerating-the-development-of-life-saving-treatments - Subject: GPT / ChatGPT / API The source text contains only a headline with no information about any AI product, model, feature, pricing, or other announcement. ### Vercel — Latency numbers every frontend developer should know - Company: Vercel (vercel.com) - Announced: 2024-04-23T13:00:00+00:00 - Category: not stated - Coverage: not counted - Announcement: no - Group: routine - Source: https://vercel.com/blog/latency-numbers-every-web-developer-should-know - Record: https://forck.live/items/20989-latency-numbers-every-frontend-developer-should-know - Subject: v0 / AI SDK / AI Gateway Vercel published a reference table of network and browser latencies relevant to frontend developers, covering metrics from WiFi (1–4ms) to worst-case intercontinental round trips (300ms). The post explains how these numbers compound in network waterfalls and notes that technologies like React Server Components can move such waterfalls to the server, where the same request pattern may be 100 times faster. ### Replicate — Run Snowflake Arctic with an API - Company: Replicate (replicate.com) - Announced: 2024-04-23 - Category: api-release - Coverage: not counted - Announcement: no - Group: routine - Source: https://replicate.com/blog/run-arctic-with-an-api - Record: https://forck.live/items/2539-run-snowflake-arctic-with-an-api - Subject: Platform Replicate announces that Snowflake Arctic, a large open-source language model with Apache 2.0 license, is now accessible via Replicate's API, providing code examples in JavaScript, Python, and cURL. ### Hugging Face — Introducing the Open Chain of Thought Leaderboard - Company: Hugging Face (huggingface.co) - Announced: 2024-04-23 - Category: developer-tool-release - Coverage: not counted - Announcement: yes - Group: announcements - Source: https://huggingface.co/blog/leaderboard-cot - Record: https://forck.live/items/1901-introducing-the-open-chain-of-thought-leaderboard - Subject: Platform Hugging Face introduces the Open Chain of Thought Leaderboard, which tracks LLMs' ability to generate effective chain-of-thought reasoning traces by measuring the accuracy gain from chain-of-thought prompting on challenging reasoning tasks. ### OpenAI — Introducing more enterprise-grade features for API customers - Company: OpenAI (openai.com) - Announced: 2024-04-23 - Category: capability-change - Coverage: not counted - Announcement: yes - Group: announcements - Source: https://openai.com/index/more-enterprise-grade-features-for-api-customers - Record: https://forck.live/items/783-introducing-more-enterprise-grade-features-for-api-customers - Subject: GPT / ChatGPT / API OpenAI announces new enterprise-grade features for API customers, including security enhancements, Assistants API updates, and cost management tools. ### LG AI Research — [LG TECH CONFERENCE 2024] Join LG in Shaping the Future of R&D - Company: LG AI Research (lgresearch.ai) - Announced: 2024-04-22 - Category: not stated - Coverage: not counted - Announcement: no - Group: routine - Source: https://www.lgresearch.ai/news/view?seq=426 - Record: https://forck.live/items/11028-lg-tech-conference-2024-join-lg-in-shaping-the-future-of-r-d - Subject: EXAONE LG AI Research participated in the LG TECH CONFERENCE 2024, held on April 4 at LG Sciencepark in Magok, which attracted over 300 Korean master's and doctoral students in science and engineering. The event included Tech Sessions, Poster Sessions, an AMA session, and a Networking Session. LG AI Research also held a pre-event, LG AI Tech CONNECT Seoul, a recruitment networking event for AI talent. ### Microsoft AI — What is AI anyway? - Company: Microsoft AI (microsoft.ai) - Announced: 2024-04-22 - Category: not stated - Coverage: not counted - Announcement: no - Group: routine - Source: https://microsoft.ai/news/what-is-ai-anyway - Record: https://forck.live/items/8441-what-is-ai-anyway - Subject: MAI Microsoft AI CEO Mustafa Suleyman proposes viewing AI as a new digital species, arguing that even those closest to its development struggle to describe where the technology is headed. The piece accompanies a Q&A with TED's head, Chris Anderson. ### LG AI Research — [LG TECH CONFERENCE 2024] Join LG in Shaping the Future of R&D - Company: LG AI Research (lgresearch.ai) - Announced: 2024-04-22 - Category: not stated - Coverage: not counted - Announcement: no - Group: routine - Source: https://www.lgresearch.ai/blog/view?seq=426 - Record: https://forck.live/items/4856-lg-tech-conference-2024-join-lg-in-shaping-the-future-of-r-d - Subject: EXAONE This is a blog post about LG TECH CONFERENCE 2024, a recruitment and networking event for R&D talent, with no new product or model announcements. ### Hugging Face — Jack of All Trades, Master of Some, a Multi-Purpose Transformer Agent - Company: Hugging Face (huggingface.co) - Announced: 2024-04-22 - Category: open-weight-release - Coverage: not counted - Announcement: yes - Group: models - Source: https://huggingface.co/blog/jat - Record: https://forck.live/items/1902-jack-of-all-trades-master-of-some-a-multi-purpose-transformer-agent - Subject: Platform Hugging Face releases Jack of All Trades (JAT), an open-source multi-modal Transformer agent capable of playing video games, controlling robots, and understanding navigation commands, along with expert RL agents and the JAT dataset.