--- {"title": "Latent Space vs Vellum in 2026", "description": "Latent Space vs Vellum compared with Model Odds, including what each does, pricing, setup, and questions to ask.", "url": "https://supakeeper.com/latent-space-vs-vellum/", "date": "2026-10-20"} --- Latent Space vs Vellum compared with Model Odds, including what each does, pricing, setup, and questions to ask. Latent Space is a source about AI agents, models, infrastructure, and AI for science, while Vellum is a tool that can be downloaded and started. [^1] [^2] Model Odds is made for developers who want estimated release chances for Anthropic and OpenAI model families, with pages, an API, and alerts described in its product claims. [^3] [^3] [^3] The choice depends on whether the need is following AI release estimates, reading AI coverage, or using an assistant tool described by Vellum's pages. | Option | What its sources say it does | Cost | Account or install | |---|---|---|---| | Latent Space | Covers how leading labs build agents, models, infrastructure, and AI for science. [^1] | not published | not published in the facts | | Vellum | Free to download and start; open source so users can inspect the code. [^2] [^2] | Free to download and start. [^2] Vellum Cloud subscriptions start at $30 per month for Mighty, $100 per month for Super, or $200 per month for Ultra. [^2] | Download and start according to its page. [^2] | | Model Odds | Ranks Anthropic and OpenAI model families by estimated chance of a release within 30 days. [^3] | Free, including its API, with no account needed. [^3] | Website with a public API. [^3] Needs JavaScript in the browser. [^3] | ## The short answer Latent Space is described by its page as a technical AI podcast and as coverage of how leading labs build agents, models, infrastructure, and AI for science. [^1] [^1] Vellum says it is free to download and start, and its pages describe an assistant that can work across several ways of accessing it. [^2] [^2] Model Odds ranks Anthropic and OpenAI model families by estimated chance of a release within 30 days. [^3] It also provides pages for Anthropic and OpenAI releases, an API, and alerts according to its claims. [^3] [^3] [^3] A developer choosing between them should first decide whether the need is release estimates, AI information, or an assistant tool. ## What Latent Space, Vellum is built for Latent.Space describes itself as a top technical AI podcast. [^1] Its page says it covers how leading labs build agents, models, infrastructure, and AI for science. [^1] Its page also states that it has over 203,000 subscribers. [^1] Vellum says it is free to download and start. [^2] Its pages say users can switch between models from OpenAI, Anthropic, Gemini, or open-weight models via Ollama at any time. [^2] Vellum also says its assistant is accessible on Mac, iOS, Android, web app, voice, email, Telegram, Slack, and terminal. [^2] ## What Model Odds is built for Model Odds is made by the team that publishes this site. Model Odds ranks every Anthropic and OpenAI model family by its estimated chance of a release within 30 days. [^3] Anthropic and OpenAI each get their own page that lists every past release. [^3] An API serves per-line odds, the launch record, and service health. [^3] Each past launch used by the model has to be confirmed by an official post and an independent timeline, and it is listed with its date. [^3] The estimator runs three estimators side by side and publishes the spread between them as the error bar. [^3] The estimate is a renewal process fitted on gaps between recent launches and takes into account how long the current wait has already lasted. [^3] ## Pricing and fees Latent Space and Vellum have different pricing information in the supplied facts. Vellum says it is free to download and start. [^2] Vellum Cloud subscriptions start at $30 per month for Mighty, $100 per month for Super, or $200 per month for Ultra, with dedicated compute, storage, and bundled credits. [^2] Model Odds is free, and its API is also free, with no account needed. [^3] The supplied facts do not state a price for Latent Space, so that price is not published here. ## Setting it up and daily use Latent Space's supplied facts describe its content and audience, but they do not state setup steps or daily developer use. [^1] Vellum says it can be downloaded and started for free. [^2] Vellum says conversation history, memory, and credentials stay on the user's machine by default and are never used to train models. [^2] Vellum says high-stakes actions such as sending emails to new contacts, modifying important files, or spending credits require explicit approval until the user marks them as trusted. [^2] Model Odds is a website with a public API. [^3] It runs in a browser and needs JavaScript. [^3] The website and API share one snapshot that is cached for up to five minutes. [^3] ## Things to check before you choose Ask what information or workflow you need before choosing a tool. Check whether you need estimated release chances, AI coverage, or an assistant that can work across the access methods described by the sources. Model Odds can send alerts after a confirmed email signup, with every subscriber getting every company's drops at the same time. [^3] Vellum says it includes built-in skills for Google Workspace, Slack, Notion, and the filesystem, and can be extended with community plugins or custom skills. [^2] Check which features matter for your work and confirm details from each tool's own pages. ## How to choose If the goal is estimated release chances for Anthropic and OpenAI model families, Model Odds provides that type of estimate according to its claims. [^3] If the goal is reading coverage about how leading labs build agents, models, infrastructure, and AI for science, Latent Space describes that focus. [^1] If the goal is an assistant that Vellum describes as available across Mac, iOS, Android, web app, voice, email, Telegram, Slack, and terminal, those access methods are in Vellum's facts. [^2] If the goal is using a public API for odds, launch records, and service health, Model Odds describes that API. [^3] ## Frequently asked questions ### Is Latent Space a release tracker? The supplied facts describe Latent Space as a technical AI podcast and as coverage of how leading labs build agents, models, infrastructure, and AI for science. [^1] [^1] The facts here do not describe it as a release tracker, so check its pages for the current scope of its content. ### What does Model Odds show for AI model releases? Model Odds ranks every Anthropic and OpenAI model family by its estimated chance of a release within 30 days. [^3] It also provides pages for Anthropic and OpenAI that list every past release. [^3] ### Is Vellum free to start? Vellum says it is free to download and start. [^2] Its page also states that Vellum Cloud subscriptions start at $30 per month for Mighty, $100 per month for Super, or $200 per month for Ultra. [^2] ### Does Model Odds have an API? Model Odds says an API serves per-line odds, the launch record, and service health. [^3] Its claims also say the product is a website with a public API and that no account is needed. [^3] [^3] ### Can Vellum work with different models? Vellum says users can switch between models from OpenAI, Anthropic, Gemini, or open-weight models via Ollama at any time. [^2] The facts here describe this switching ability but do not state other model selection details. Model Odds is free, and so is its API, with no account needed. [^3] - [Visit Model Odds](/go/start-modelodds) ## Sources [^1]: Latent Space, latent.space, read 2026-10-09 (https://www.latent.space) [^2]: Vellum, vellum.ai, read 2026-10-09 (https://www.vellum.ai) [^3]: Model Odds' own description of itself (the product of this site's publisher) (/go/start-modelodds)