ChatGPT images 2.0 🎨, Qwen3.5-Omni 🧠, always-on ChatGPT agents 🤖

OpenAI introduced an upgraded image model with improved text
rendering, multi-image reasoning, and higher fidelity outputs,
enabling complex
assets ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌  ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ 

 Sign Up [1] |Advertise [2]|View Online [3]

		TLDR

		TOGETHER WITH [WorkOS] [4]

TLDR AI 2026-04-22

 NPX WORKOS: FROM AUTH INTEGRATION TO ENVIRONMENT MANAGEMENT, ZERO
CLICKOPS (SPONSOR) [4]

 NPX WORKOS@LATEST launches an AI agent, powered by Claude [5], that
reads your project, detects your framework, and writes a complete auth
integration into your codebase. No signup required. It creates an
environment, populates your keys, and you claim your account later
when you're ready.

But the CLI goes way beyond installation. [6] WorkOS Skills make your
coding agent a WorkOS expert. WORKOS SEED defines your environment as
code. WORKOS DOCTOR finds and fixes misconfigurations. And once you're
authenticated, your agent can manage users, orgs, and environments
directly from the terminal. No more ClickOps.

See how it works → [4]

🚀

HEADLINES & LAUNCHES

 CHATGPT IMAGES 2.0 (6 MINUTE READ) [7]

 OpenAI introduced an upgraded image model with improved text
rendering, multi-image reasoning, and higher fidelity outputs,
enabling complex assets like comics and marketing visuals.

 OPENAI DEVELOPS PLATFORM FOR ALWAYS-ON AGENTS ON CHATGPT (2 MINUTE
READ) [8]

 OpenAI is developing an always-on agent platform within ChatGPT,
codenamed Hermes, that allows users to create and continuously run
custom agents. This platform includes features for creating workflows,
integrating skills, and scheduling tasks, enabling agents to act
independently rather than waiting for prompts. OpenAI's move presents
strong competition to existing platforms like Notion by bringing such
capabilities to a vast user base.

 QWEN3.5-OMNI TECHNICAL REPORT (4 MINUTE READ) [9]

 Qwen3.5-Omni is a large-scale multimodal model with hundreds of
billions of parameters that natively processes text, audio, images,
and video within a unified architecture. The model supports a 256k
token context length to seamlessly handle up to 10 hours of audio or
400 seconds of high definition video in real time. It leverages a
Hybrid Attention Mixture of Experts framework alongside a dynamic
alignment technique called ARIA to generate highly stable and
emotionally nuanced multilingual speech synthesis with minimal
latency.

🧠

DEEP DIVES & ANALYSIS

 IMAGE GENERATION PROMPTING GUIDE (38 MINUTE READ) [10]

 A practical guide that outlines prompting strategies for image
generation, covering techniques for controlling style, structure, and
fidelity in production image workflows.

 CODING AGENTS IGNORE THEIR OWN BUDGETS (5 MINUTE READ) [11]

 Ramp Labs discovered that autonomous coding agents completely ignore
passive token limits and cannot reliably regulate their own spending.
When forced to explicitly approve or deny budget extensions, the
models exhibited severe self-attribution bias by overly praising their
own progress and nearly always approving more spend. To effectively
manage costs, researchers had to separate the working agent from
financial decisions by deploying an independent controller model that
evaluates objective workspace snapshots.

 WHEN CAN LLMS LEARN TO REASON WITH WEAK SUPERVISION? (4 MINUTE READ)
[12]

 This study found that models with extended pre-saturation phases
generalize well from minimal examples and tolerate noise, while
rapidly saturating models fail. The key issue is unfaithful reasoning,
where models memorize answers rather than learning transferable
reasoning. Continual pre-training and supervised fine-tuning on
explicit reasoning traces improve reasoning faithfulness and
generalization under weak supervision.

🧑‍💻

ENGINEERING & RESEARCH

 GOOGLE CLOUD NEXT STARTS TODAY! (SPONSOR) [13]

 If you're building AI applications, you need infrastructure that can
actually handle the compute.

Google uses Tensor Processing Units (TPUs) - custom-built hardware
accelerators designed specifically for large-scale AI workloads. It's
the exact same accelerator system powering Gemini and powers billions
of user requests across Search and Maps.

Ready to learn how to leverage TPUs for your own training and
inference workloads?

Start the course → [14]

 CRABTRAP: AN LLM-AS-A-JUDGE HTTP PROXY TO SECURE AGENTS IN PRODUCTION
(9 MINUTE READ) [15]

 CrabTrap is an open-source HTTP/HTTPS proxy that intercepts every
request an AI agent makes and uses LLM-as-a-judge to determine if the
request matches a policy of allowed traffic for that agent. Agents
need real credentials, but can hallucinate destructive actions or get
prompt-injected. This can have production consequences. CrabTrap
introduces guardrails that represent a meaningful step forward in the
security of agent harnesses in production environments.

 STITCH'S DESIGN.MD FORMAT IS NOW OPEN-SOURCE SO YOU CAN USE IT ACROSS
PLATFORMS. (1 MINUTE READ) [16]

 Stitch's DESIGN.md lets users export or import design rules from
project to project. Stitch understands the reasoning behind design
systems and can generate user interfaces that match branches. Google
has open sourced the draft specification for DESIGN.md, which can be
used across any tool or platform. A video breaking down the format is
available in the article.

 CRITICAL BITS IN NEURAL NETWORKS (6 MINUTE READ) [17]

 Deep Neural Lesion (DNL) identifies highly sensitive parameters where
flipping just a few bits can collapse model performance across vision
and language tasks. The work also shows that protecting a small subset
of these bits can mitigate such failures.

🎁

MISCELLANEOUS

 OPENAI IS WORKING WITH CONSULTANTS TO SELL CODEX (3 MINUTE READ) [18]

 OpenAI is working with several consulting firms to help sell its AI
coding tool Codex to businesses. Codex now has four million weekly
active users, up from three million just two weeks ago. The Codex
consulting program is part of OpenAI's push to focus on coding and
enterprise businesses. Consulting partners will get access to an AI
coding tool as part of the program.

 SAM ALTMAN THROWS SHADE AT ANTHROPIC'S CYBER MODEL, MYTHOS:
‘FEAR-BASED MARKETING' (2 MINUTE READ) [19]

 OpenAI CEO Sam Altman called out Anthropic's new cybersecurity model
during a podcast appearance this week, saying the company was using
fear to make its product sound more impressive than it actually is.
Anthropic announced its Mythos model earlier this month and only
released it to a small cohort of enterprise customers with the claim
that the model was too powerful to be released to the public as
cybercriminals would weaponize it. Altman said that Anthropic's
fear-based marketing was a good way to keep AI in the hands of a small
and exclusive elite. Fear-based marketing is prevalent in the AI
industry, and it has also come from Altman himself.

⚡

QUICK LINKS

 BUILD, DEPLOY, AND SCALE AI INFRASTRUCTURE FASTER WITH RUNPOD
(SPONSOR) [20]

 Runpod is a GPU cloud developers use to launch pods, run inference,
and autoscale on demand. Pay only for what you use. Start scaling
today. [21]

 ANTHROPICS WORKS ON ITS ALWAYS-ON AGENT WITH UI EXTENSIONS (3 MINUTE
READ) [22]

 Anthropic's "Conway" is an always-on agent with UI extensions
available on web and mobile, allowing users to manage connectors,
install extensions, and configure the environment.

 DEEP RESEARCH MAX: A STEP CHANGE FOR AUTONOMOUS RESEARCH AGENTS (6
MINUTE READ) [23]

 Google has introduced Deep Research and Deep Research Max, leveraging
the Gemini 3.1 Pro model to enhance autonomous research capabilities.

 TLDR IS HIRING A CURATOR FOR TLDR AI (3-5 HRS/WEEK, FULLY REMOTE)
[24]

 We're hiring an engineer/researcher at a major AI lab or startup to
help write for 1M+ subscribers. Curators have been invited to Google
I/O and OpenAI DevDay, scouted for Tier 1 VCs, and get early access to
unreleased TLDR products. Learn more [25].

 THE FALL OF THE THEOREM ECONOMY (63 MINUTE READ) [26]

 It will eventually become unthinkable to do math without AI
assistance, just like it has become unthinkable to do math without set
theory and LaTeX.

 AGENT WORLD TRAINING ARENA (3 MINUTE READ) [27]

 Agent-World describes a self-evolving environment that generates
tasks and feedback loops to continuously train and improve autonomous
agents.

Want to advertise in TLDR? 📰

 If your company is interested in reaching an audience of AI
professionals and decision makers, you may want to ADVERTISE WITH US
[28].

Want to work at TLDR? 💼

 APPLY HERE [29], CREATE YOUR OWN ROLE [30] or send a friend's resume
to jobs@tldr.tech and get $1k if we hire them! TLDR is one of INC.'S
BEST BOOTSTRAPPED BUSINESSES [31] of 2025.

 If you have any comments or feedback, just respond to this email!

Thanks for reading,
Andrew Tan [32], Ali Aminian [33], & Jacob Turner [34]

 Manage your subscriptions [35] to our other newsletters on tech,
startups, and programming. Or if TLDR AI isn't for you, please
unsubscribe [36].

Links:
------
[1] https://tldr.tech/ai
[2] https://advertise.tldr.tech/
[3] https://a.tldrnewsletter.com/web-version?ep=1&lc=97c90d72-3e48-11f1-ab0b-7fc89b18583c&p=5ce7f8f4-3e2e-11f1-b7ea-0524d2d55bb4&pt=campaign&t=1776864809&s=1f36eccf2f330ebcdb324bfdc4ea95b867f0f21633f420fcbb1c59eb726b6e14
[4] https://workos.com/docs/authkit/cli-installer
[5] https://links.tldrnewsletter.com/YNVjJO
[6] https://workos.com/blog/agent-experience
[7] https://links.tldrnewsletter.com/mCuG8v
[8] https://www.testingcatalog.com/openai-develops-platform-for-always-on-agents-on-chatgpt/
[9] https://www.alphaxiv.org/abs/2604.15804
[10] https://developers.openai.com/cookbook/examples/multimodal/image-gen-models-prompting-guide
[11] https://links.tldrnewsletter.com/sktuqW
[12] https://salmanrahman.net/rlvr-weak-supervision
[13] https://www.google.com/url?sa=j&url=https%3A%2F%2Fwww.googlecloudevents.com%2Fnext-vegas%2F%3Futm_source%3Dgoogle%26utm_medium%3Dgoogle%26utm_campaign%3DFY26-Q2-GLOBAL-GLO27877-physicalevent-er-next26-mc-105752%26utm_content%3Dgoogle-sem-keywords%26utm_term%3D-%26gclsrc%3Daw.ds%26gad_source%3D1%26gad_campaignid%3D23326881109%26gbraid%3D0AAAAApdQcwfxC0NmS3Yma9d70r0v82u6J%26gclid%3DCjwKCAjwwJzPBhBREiwAJfHRnRe0QMj35LOFkFZz7KiAhZTdSwbGm4qUtAuEFgJIl6kgJPHmuF_tuxoC9jYQAvD_BwE&uct=1762650427&usg=mYkjYgezYTU6arfmTcJs5iV-CyA.&opi=73833047&source=chat
[14] https://www.skills.google/paths/2806/course_templates/1405
[15] https://links.tldrnewsletter.com/K4dyDN
[16] https://blog.google/innovation-and-ai/models-and-research/google-labs/stitch-design-md/
[17] https://mkimhi.github.io/DNL/
[18] https://links.tldrnewsletter.com/hbrDi2
[19] https://techcrunch.com/2026/04/21/sam-altman-throws-shade-at-anthropics-cyber-model-mythos-fear-based-marketing/
[20] https://www.runpod.io/?inflect=&targetid=kwd-461794446387&adgroupid=189724568342&loc_interest=&loc_physical=9012242&hsa_acc=4558579452&hsa_cam=23516214656&hsa_grp=189724568342&hsa_ad=797799116312&hsa_src=g&hsa_tgt=kwd-461794446387&hsa_kw=runpod&hsa_mt=e&hsa_net=adwords&hsa_ver=3&gad_source=1&gad_campaignid=23516214656&gbraid=0AAAAAoZSBmm_mSAwIfhLEqmrXZN5buVzH
[21] https://fandf.co/4vN7sBQ
[22] https://www.testingcatalog.com/anthropics-works-on-its-always-on-agent-with-new-ui-extensions/
[23] https://blog.google/innovation-and-ai/models-and-research/gemini-models/next-generation-gemini-deep-research
[24] https://jobs.ashbyhq.com/tldr.tech/038c4419-5b48-4279-a75e-6f7a0afdb240
[25] https://jobs.ashbyhq.com/tldr.tech/038c4419-5b48-4279-a75e-6f7a0afdb240
[26] https://davidbessis.substack.com/p/the-fall-of-the-theorem-economy
[27] https://agent-tars-world.github.io/-/

🚀

HEADLINES & LAUNCHES

 CHATGPT IMAGES 2.0

 OpenAI introduced an upgraded image model with improved text rendering, multi-image reasoning, and higher fidelity outputs, enabling complex assets like comics and marketing visuals.

 OPENAI DEVELOPS PLATFORM FOR ALWAYS-ON AGENTS ON CHATGPT

 OpenAI is developing an always-on agent platform within ChatGPT, codenamed Hermes, that allows users to create and continuously run custom agents. This platform includes features for creating workflows, integrating skills, and scheduling tasks, enabling agents to act independently rather than waiting for prompts. OpenAI's move presents strong competition to existing platforms like Notion by bringing such capabilities to a vast user base.

 QWEN3.5-OMNI TECHNICAL REPORT

 Qwen3.5-Omni is a large-scale multimodal model with hundreds of billions of parameters that natively processes text, audio, images, and video within a unified architecture. The model supports a 256k token context length to seamlessly handle up to 10 hours of audio or 400 seconds of high definition video in real time. It leverages a Hybrid Attention Mixture of Experts framework alongside a dynamic alignment technique called ARIA to generate highly stable and emotionally nuanced multilingual speech synthesis with minimal latency.

🧠

DEEP DIVES & ANALYSIS

 IMAGE GENERATION PROMPTING GUIDE

 A practical guide that outlines prompting strategies for image generation, covering techniques for controlling style, structure, and fidelity in production image workflows.

 CODING AGENTS IGNORE THEIR OWN BUDGETS

 Ramp Labs discovered that autonomous coding agents completely ignore passive token limits and cannot reliably regulate their own spending. When forced to explicitly approve or deny budget extensions, the models exhibited severe self-attribution bias by overly praising their own progress and nearly always approving more spend. To effectively manage costs, researchers had to separate the working agent from financial decisions by deploying an independent controller model that evaluates objective workspace snapshots.

 WHEN CAN LLMS LEARN TO REASON WITH WEAK SUPERVISION?

 This study found that models with extended pre-saturation phases generalize well from minimal examples and tolerate noise, while rapidly saturating models fail. The key issue is unfaithful reasoning, where models memorize answers rather than learning transferable reasoning. Continual pre-training and supervised fine-tuning on explicit reasoning traces improve reasoning faithfulness and generalization under weak supervision.

🧑‍💻

ENGINEERING & RESEARCH

 CRABTRAP: AN LLM-AS-A-JUDGE HTTP PROXY TO SECURE AGENTS IN PRODUCTION

 CrabTrap is an open-source HTTP/HTTPS proxy that intercepts every request an AI agent makes and uses LLM-as-a-judge to determine if the request matches a policy of allowed traffic for that agent. Agents need real credentials, but can hallucinate destructive actions or get prompt-injected. This can have production consequences. CrabTrap introduces guardrails that represent a meaningful step forward in the security of agent harnesses in production environments.

 STITCH'S DESIGN.MD FORMAT IS NOW OPEN-SOURCE SO YOU CAN USE IT ACROSS PLATFORMS

 Stitch's DESIGN.md lets users export or import design rules from project to project. Stitch understands the reasoning behind design systems and can generate user interfaces that match branches. Google has open sourced the draft specification for DESIGN.md, which can be used across any tool or platform.

 CRITICAL BITS IN NEURAL NETWORKS

 Deep Neural Lesion (DNL) identifies highly sensitive parameters where flipping just a few bits can collapse model performance across vision and language tasks. The work also shows that protecting a small subset of these bits can mitigate such failures.

🎁

MISCELLANEOUS

 OPENAI IS WORKING WITH CONSULTANTS TO SELL CODEX

 OpenAI is working with several consulting firms to help sell its AI coding tool Codex to businesses. Codex now has four million weekly active users, up from three million just two weeks ago. The Codex consulting program is part of OpenAI's push to focus on coding and enterprise businesses.

 SAM ALTMAN THROWS SHADE AT ANTHROPIC'S CYBER MODEL, MYTHOS: 'FEAR-BASED MARKETING'

 OpenAI CEO Sam Altman called out Anthropic's new cybersecurity model during a podcast appearance this week, saying the company was using fear to make its product sound more impressive than it actually is. Anthropic announced its Mythos model and only released it to a small cohort of enterprise customers with the claim that the model was too powerful to be released to the public as cybercriminals would weaponize it. Altman said that Anthropic's fear-based marketing was a good way to keep AI in the hands of a small and exclusive elite.

⚡

QUICK LINKS

 ANTHROPICS WORKS ON ITS ALWAYS-ON AGENT WITH UI EXTENSIONS

 Anthropic's "Conway" is an always-on agent with UI extensions available on web and mobile, allowing users to manage connectors, install extensions, and configure the environment.

 DEEP RESEARCH MAX: A STEP CHANGE FOR AUTONOMOUS RESEARCH AGENTS

 Google has introduced Deep Research and Deep Research Max, leveraging the Gemini 3.1 Pro model to enhance autonomous research capabilities.

 THE FALL OF THE THEOREM ECONOMY

 It will eventually become unthinkable to do math without AI assistance, just like it has become unthinkable to do math without set theory and LaTeX.

 AGENT WORLD TRAINING ARENA

 Agent-World describes a self-evolving environment that generates tasks and feedback loops to continuously train and improve autonomous agents.

Tu es l'extracteur de contenu de Breviat. On te fournit le contenu Markdown nettoyé d'une newsletter.

Ta mission : produire une version PROPRE du contenu en supprimant tout ce qui n'est pas de l'information utile au lecteur. Tu es un FILTRE, pas un résumeur.

À RETIRER :
- Publicités, encarts sponsors, mentions "sponsorisé par X", "ad", "présenté par"
- Intros vides : formules de bienvenue, météo de l'humeur de l'auteur, anecdotes personnelles non liées au contenu
- Appels à l'action marketing : s'abonner à la newsletter, parrainer un ami, "follow us on Twitter", "join our Discord"
- Signatures, mentions légales, adresses postales, "view in browser", "unsubscribe"
- Boutons / CTAs / "cliquez ici" / "lire la suite" sans contenu derrière
- Promotions d'autres produits / événements / formations payantes de l'auteur ou de tiers
- Encarts récurrents type "Read of the day" ou "Quote of the day" sans valeur informationnelle propre

À CONSERVER (intégralement, sans résumer ni reformuler) :
- Toutes les annonces, news, analyses, commentaires factuels
- Les chiffres, dates, noms d'entreprises, citations
- Les explications techniques
- Les liens vers des sources réelles (annonces officielles, papers, articles cités)
- La structure (titres, sous-titres, listes)

RÈGLES :
- Ne reformule pas. Garde la formulation d'origine.
- Ne résume pas, ne condense pas. Si une section fait 200 mots et est utile, garde 200 mots.
- N'ajoute aucun contenu (pas de titres ni de transitions de ton cru).
- Ne fabrique aucune URL. Garde celles d'origine, ou retire-les.
- Si la newsletter entière est de la pub / promo / contenu inutile, sors un Markdown vide (rien d'autre).

Sortie : UNIQUEMENT le Markdown nettoyé, sans préambule ni commentaire sur ton travail.