<!DOCTYPE html><html lang="en"><head><meta http-equiv="Content-Type" content="text/html charset=UTF-8"><meta charset="UTF-8"><meta name="viewport" content="width=device-width"><meta name="x-apple-disable-message-reformatting"><title>TLDR AI</title><meta name="color-scheme" content="light dark"><meta name="supported-color-schemes" content="light dark"><style type="text/css">
:root {
color-scheme: light dark; supported-color-schemes: light dark;
}
*,
*:after,
*:before {
-webkit-box-sizing: border-box; -moz-box-sizing: border-box; box-sizing: border-box;
}
* {
-ms-text-size-adjust: 100%; -webkit-text-size-adjust: 100%;
}
html,
body,
.document {
width: 100% !important; height: 100% !important; margin: 0; padding: 0;
}
body {
-webkit-font-smoothing: antialiased; -moz-osx-font-smoothing: grayscale; text-rendering: optimizeLegibility;
}
div[style*="margin: 16px 0"] {
margin: 0 !important;
}
table,
td {
mso-table-lspace: 0pt; mso-table-rspace: 0pt;
}
table {
border-spacing: 0; border-collapse: collapse; table-layout: fixed; margin: 0 auto;
}
img {
-ms-interpolation-mode: bicubic; max-width: 100%; border: 0;
}
*[x-apple-data-detectors] {
color: inherit !important; text-decoration: none !important;
}
.x-gmail-data-detectors,
.x-gmail-data-detectors *,
.aBn {
border-bottom: 0 !important; cursor: default !important;
}
.btn {
-webkit-transition: all 200ms ease; transition: all 200ms ease;
}
.btn:hover {
background-color: #f67575; border-color: #f67575;
}
* {
font-family: Arial, Helvetica, sans-serif; font-size: 18px;
}
@media screen and (max-width: 600px) {
.container {
width: 100%; margin: auto;
}
.stack {
display: block!important; width: 100%!important; max-width: 100%!important;
}
.btn {
display: block; width: 100%; text-align: center;
}
}
body,
p,
td,
tr,
.body,
table,
h1,
h2,
h3,
h4,
h5,
h6,
div,
span {
background-color: #FEFEFE !important; color: #010101 !important;
}
@media (prefers-color-scheme: dark) {
body,
p,
td,
tr,
.body,
table,
h1,
h2,
h3,
h4,
h5,
h6,
div,
span {
background-color: #27292D !important; color: #FEFEFE !important;
}
}
a {
color: inherit !important; text-decoration: underline !important;
}
</style><!--[if mso | ie]>
<style type="text/css">
a {
background-color: #FEFEFE !important; color: #010101 !important;
}
@media (prefers-color-scheme: dark) {
a {
background-color: #27292D !important; color: #FEFEFE !important;
}
}
</style>
<![endif]--></head><body class="">
<div style="display: none; max-height: 0px; overflow: hidden;">Google's Gemini 3.1 Flash TTS enhances text-to-speech with improved expressivity and controllability, featuring a notable Elo score of 1,211 </div>
<div style="display: none; max-height: 0px; overflow: hidden;">
<br>
</div>
<table align="center" class="document"><tbody><tr><td valign="top">
<table align="center" border="0" cellpadding="0" cellspacing="0" class="container" width="600"><tbody><tr class="inner-body"><td>
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr class="header"><td bgcolor="" class="container">
<table width="100%"><tbody><tr><td class="container">
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" style="margin-top: 0px;" width="100%"><tbody><tr><td style="padding: 0px;">
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div style="text-align: center;">
<span style="margin-right: 0px;"><a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Ftldr.tech%2Fai%3Futm_source=tldrai/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/kRMstzatpRze7ELBYZRNl3v7Pod6CnCtzPSeV2eNZM0=452" rel="noopener noreferrer" target="_blank"><span>Sign Up</span></a>
|<span style="margin-right: 2px; margin-left: 2px;"><a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fadvertise.tldr.tech%2F%3Futm_source=tldrai%26utm_medium=newsletter%26utm_campaign=advertisetopnav/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/eH2tMcOAUvXVbLF8BKSFW9BOuoywREWknptGBo1ECLc=452" rel="noopener noreferrer" target="_blank"><span>Advertise</span></a></span>|<span style="margin-left: 2px;"><a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fa.tldrnewsletter.com%2Fweb-version%3Fep=1%26lc=1571a9f6-84b7-11f0-b464-47c5c04ad337%26p=b97b2e02-3978-11f1-85a8-e709599e4e50%26pt=campaign%26t=1776345636%26s=82d74555ecf199cd86331bc48f847dfeb8f0c23a3577d7fe02fe55bf4ca5f63c/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/7G-SMEGEzJEvPdOPRqIO_hYnl2hmsMgnH9UQopef2ic=452"><span>View Online</span></a></span>
<br>
</span></div>
</td></tr></tbody></table>
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="text-align: center;"><span data-darkreader-inline-color="" style="--darkreader-inline-color:#3db3ff; color: rgb(51, 175, 255) !important; font-size: 30px;">T</span><span style="font-size: 30px;"><span data-darkreader-inline-color="" style="color: rgb(232, 192, 96) !important; --darkreader-inline-color:#e8c163; font-size:30px;">L</span><span data-darkreader-inline-color="" style="color: rgb(101, 195, 173) !important; --darkreader-inline-color:#6ec7b2; font-size:30px;">D</span></span><span data-darkreader-inline-color="" style="--darkreader-inline-color:#dd6e6e; color: rgb(220, 107, 107) !important; font-size: 30px;">R</span>
<br>
</td></tr></tbody></table>
<br>
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr id="together-with"><td align="center" height="20" style="vertical-align:middle !important;" valign="middle" width="100%"><strong style="vertical-align:middle !important; height: 100%;">Together With </strong>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fmetronome.com%2Fblog%2F2026-trends-from-cataloging-50-ai-pricing-models%3Futm_campaign=index%26utm_medium=newsletter%26utm_source=tldr-ai%26utm_content=/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/RHDtt7d5nRd7aKyTrtTaPapbvZ64mdW8u06xfcFWi2g=452"><img src="https://images.tldr.tech/metronome2.png" valign="middle" style="vertical-align: middle !important; height: 100%;" alt="Metronome"></a></td></tr></tbody></table>
<table style="table-layout: fixed; width:100%;" width="100%"><tbody><tr><td style="padding:0;border-collapse:collapse;border-spacing:0;margin:0;">
<div style="text-align: center;">
<h1><strong>TLDR AI <span id="date">2026-04-16</span></strong></h1>
</div>
</td></tr></tbody></table>
<table style="table-layout: fixed; width:100%;" width="100%"><tbody><tr id="sponsy-copy"><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fmetronome.com%2Fblog%2F2026-trends-from-cataloging-50-ai-pricing-models%3Futm_campaign=index%26utm_medium=newsletter%26utm_source=tldr-ai%26utm_content=/2/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/zrLHH5e6C98Eu8-pZl_XorfAGBKTU7rt1EZZzte77U0=452">
<span>
<strong>What 50+ AI pricing models reveal about AI monetization in 2026 (Sponsor)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Usage, credits, access, GBB… which monetization model makes sense for AI? Metronome's Pricing Index covers 50+ AI companies, and they've shared some takeaways in <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fmetronome.com%2Fblog%2F2026-trends-from-cataloging-50-ai-pricing-models%3Futm_campaign=index%26utm_medium=newsletter%26utm_source=tldr-ai%26utm_content=/3/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/jsafXUmgwIfMlJca0v9ybHx3KNFQOp1dfWxq-shaQBI=452" rel="noopener noreferrer nofollow" target="_blank"><span>a new blog</span></a>.
<p></p>
<p>Read it to see:</p>
<p>→ Why hybrid pricing has become the norm</p>
<p>→ How companies like <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fmetronome.com%2Fblog%2F2026-trends-from-cataloging-50-ai-pricing-models%3Futm_campaign=index%26utm_medium=newsletter%26utm_source=tldr-ai%26utm_content=/4/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/VQtoCmP9SfOxqsgdaXiFz8dv2zFSQuqpH6T9e7dzDQE=452" rel="noopener noreferrer nofollow" target="_blank"><span>ElevenLabs, Clay, and Perplexity</span></a> are handling AI credits</p>
<p>→ How split pricing schemes for consumers & developers reflect value creation</p>
<p>→ The 5 key consideration monetization leaders should take into account</p>
<p><a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fmetronome.com%2Fblog%2F2026-trends-from-cataloging-50-ai-pricing-models%3Futm_campaign=index%26utm_medium=newsletter%26utm_source=tldr-ai%26utm_content=/5/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/p4tIksaHuZNXJo-7XThI70V1aWJz0r0k9xpede6ekwI=452" rel="noopener noreferrer nofollow" target="_blank"><span>Read the blog</span></a>
</p>
</span></span></div>
</td></tr></tbody></table>
</td></tr></tbody></table>
</td></tr></tbody></table>
</td></tr>
<tr bgcolor=""><td class="container">
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td style="padding: 0px;">
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding-top: 0px; padding-bottom: 0px;">
<div class="text-block">
<div style="text-align: center;"><span style="font-size: 36px;"><span style="font-size:36px;">🚀</span></span></div></div>
</td></tr></tbody></table>
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding-top: 0px; padding-bottom: 0px;">
<div class="text-block">
<div style="text-align: center;">
<h1><strong>Headlines & Launches</strong></h1>
</div>
</div>
</td></tr></tbody></table>
<table style="table-layout: fixed; width: 100%;" width="100%"><tbody><tr><td style="padding:0;border-collapse:collapse;border-spacing:0;margin:0;" valign="top">
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fblog.google%2Finnovation-and-ai%2Fmodels-and-research%2Fgemini-models%2Fgemini-3-1-flash-tts%2F%3Futm_source=tldrai/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/GPzNnlL07p6LMUw7Erl-3C6hMyzR8OUBYdgbsB7EFVA=452">
<span>
<strong>Gemini 3.1 Flash TTS: the next generation of expressive AI speech (4 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Google's Gemini 3.1 Flash TTS enhances text-to-speech with improved expressivity and controllability, featuring a notable Elo score of 1,211 on the Artificial Analysis TTS leaderboard. The model supports over 70 languages and introduces audio tags for granular control of vocal style, allowing easy manipulation via natural language commands. All generated audio is watermarked with SynthID to ensure authentic content, preventing misinformation.
</span>
</span>
</div>
</td></tr></tbody></table>
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Flinks.tldrnewsletter.com%2FALGo3b/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/1Ia8e0s2LgOiZHbyKiOhv889XSGDNcDYfNHGHUoieY0=452">
<span>
<strong>OpenAI's Updated Agents SDK (4 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
OpenAI introduced updates to its Agents SDK, adding a model-native harness for cross-file and tool workflows along with sandboxed execution for safer task handling.
</span>
</span>
</div>
</td></tr></tbody></table>
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fwww.testingcatalog.com%2Fhumwork-a2p-marketplace-connects-ai-agents-with-experts%2F%3Futm_source=tldrai/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/2787RZ9AMOCZMgmclKOl9sQGdJdnz76WuXtugev9iXA=452">
<span>
<strong>Humwork A2P marketplace connects AI agents with experts (2 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Humwork launches the first Agent-to-Person (A2P) marketplace to connect AI agents with verified human experts when AI tools encounter challenges. The platform integrates with AI-centric tools like Claude Code and Replit, allowing handoffs to occur in under 30 seconds with full session context shared securely. With more than 1,000 experts available globally, Humwork boasts an 87% resolution rate and is backed by Y Combinator's P26 batch.
</span>
</span>
</div>
</td></tr></tbody></table>
</td></tr></tbody></table>
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding-top: 0px; padding-bottom: 0px;">
<div class="text-block">
<div style="text-align: center;"><span style="font-size: 36px;">🧠</span></div>
</div>
</td></tr></tbody></table>
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding-top: 0px; padding-bottom: 0px;">
<div class="text-block">
<div style="text-align: center;">
<h1><strong>Deep Dives & Analysis</strong></h1>
</div>
</div>
</td></tr></tbody></table>
<table style="table-layout: fixed; width: 100%;" width="100%"><tbody><tr><td style="padding:0;border-collapse:collapse;border-spacing:0;margin:0;" valign="top">
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fhuggingface.co%2Fblog%2Fibm-research%2Fvakra-benchmark-analysis%3Futm_source=tldrai/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/btQSEafUD8BKQDtFzfIj2AxxL39yKvuFS6zRxki--wk=452">
<span>
<strong>Evaluating Agent Reasoning (28 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
IBM Research uses an executable benchmark with thousands of APIs and documents to test multi-step agent reasoning and tool use, revealing consistent performance gaps and common failure modes.
</span>
</span>
</div>
</td></tr></tbody></table>
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fallenai.org%2Fblog%2Fevaluating-scientific-discovery-agents%3Futm_source=tldrai/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/aoo1HSOtrKrb8zg0Qb-bVS7C61V6OScAfuqYHOYUmJ8=452">
<span>
<strong>Evaluating agents for scientific discovery (7 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Many teams are claiming extraordinary things about their agents. The evidence behind these claims is usually disappointing. ScienceWorld and DiscoveryWorld are benchmarks developed to test whether AI agents can actually do science. ScienceWorld asks whether agents can 're-make' classic scientific discoveries at roughly an elementary school level, while DiscoveryWorld tests open-ended discovery at a college or PhD level. These benchmarks, open and freely available, help test what science agents are actually capable of.
</span>
</span>
</div>
</td></tr></tbody></table>
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Flinks.tldrnewsletter.com%2FsS4VPu/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/RYfNSTRNUOPNvyXX57_a_xNeApQ2wg82Woqt6WqVE5A=452">
<span>
<strong>Why do dLLMs tend to collapse in RL (3 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Diffusion Language Models (dLLMs) experience training collapse during Reinforcement Learning because their log-likelihood must be estimated using high-variance Monte Carlo sampling, which creates noisy importance ratios. These noisy ratios induce gradient spikes that push policy drift in a positive feedback loop, a problem that traditional AR methods like conditional clipping fail to solve. The newly proposed StableDRL framework stabilizes the update process by combining unconditional clipping to suppress extreme values with self-normalization tied to the effective information in the batch.
</span>
</span>
</div>
</td></tr></tbody></table>
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fblogs.nvidia.com%2Fblog%2Flowest-token-cost-ai-factories%2F%3Futm_source=tldrai/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/nneG1yEUUF-VNoOJZ9TnktvCgHbHCm72YXBgIvAt9AY=452">
<span>
<strong>Rethinking AI TCO: Why Cost per Token Is the Only Metric That Matters (6 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Cost per token is crucial for AI infrastructure TCO assessment due to its focus on delivered intelligence, integrating hardware, software, and utilization efficiencies. Unlike traditional metrics like compute cost or FLOPS per dollar, cost per token highlights real-world performance, enabling profitable AI scaling. Evidence from NVIDIA shows its Blackwell platform drastically reduces cost per token compared to Hopper, offering significant business value.
</span>
</span>
</div>
</td></tr></tbody></table>
</td></tr></tbody></table>
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding-top: 0px; padding-bottom: 0px;">
<div class="text-block">
<div style="text-align: center;"><span style="font-size: 36px;">🧑💻</span></div>
</div>
</td></tr></tbody></table>
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding-top: 0px; padding-bottom: 0px;">
<div class="text-block">
<div style="text-align: center;">
<h1><strong>Engineering & Research</strong></h1>
</div>
</div>
</td></tr></tbody></table>
<table style="table-layout: fixed; width: 100%;" width="100%"><tbody><tr><td style="padding:0;border-collapse:collapse;border-spacing:0;margin:0;" valign="top">
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Ffandf.co%2F4bvUDne%3Futm_source=tldrai/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/Jk74ZHmUZmJ70IawP_3Ka44RactAKN7Ufgd7mwy401I=452">
<span>
<strong>Run AI agents without exposing your infrastructure (Sponsor)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
You wouldn't let an unknown human into your infrastructure. Why let an unknown agent? <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Ffandf.co%2F4bvUDne/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/x87UR9fOZpU_uab06lpnXdpi2SgmAHoSBzlTnWG3DGA=452" rel="noopener noreferrer nofollow" target="_blank"><span>Teleport Beams</span></a> runs each agent in an isolated Firecracker VM with built-in identity — connected to your infrastructure and inference services with no secrets and no IAM wrestling. Zero standing privileges. Fully auditable. <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Ffandf.co%2F4bvUDne/2/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/VVW17H1QD54EtEbJ22QNb1R72IAjSyr00cHCdkD_J9M=452" rel="noopener noreferrer nofollow" target="_blank"><span>Get early access</span></a>.
</span>
</span>
</div>
</td></tr></tbody></table>
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fwww.together.ai%2Fblog%2Fparcae%3Futm_source=tldrai/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/QMkFtvQr42Fu4ixdqMjb0OLufg70Ktxj8jHK9WoBWSM=452">
<span>
<strong>Parcae: Doing more with fewer parameters using stable looped models (6 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Parcae is one of the first stable architectures for looped language models. It achieves the quality of a Transformer twice the size with clean, predictable training. Parcae increases the recurrence rather than purely scaling data, creating a new medium to scale quality. The name Parcae is a homage to the three Roman fates: Nona, Decima, and Morta.
</span>
</span>
</div>
</td></tr></tbody></table>
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Farxiv.org%2Fabs%2F2604.13036%3Futm_source=tldrai/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/aOXbGFAbn1hOQqn8W71F1PARTywPY5kCGkBawGPE9Ug=452">
<span>
<strong>NVIDIA's Lyra 2 (32 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Lyra 2.0 is a framework for generating long, camera-controlled videos that maintain 3D consistency, using geometry-guided retrieval to prevent spatial forgetting and self-augmented training to reduce temporal drift.
</span>
</span>
</div>
</td></tr></tbody></table>
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Farxiv.org%2Fabs%2F2604.09443%3Futm_source=tldrai/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/OwPdA5EQVLnzplN2Z6cWLtS2WtK_O9ojy8HU3yHoxOI=452">
<span>
<strong>Many-Tier Instruction Hierarchy in LLM Agents (1 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Researchers propose a Many-Tier Instruction Hierarchy (ManyIH) to address instruction conflicts in LLM agents, surpassing traditional models with fixed privilege levels. They introduce ManyIH-Bench, assessing models across 12 privilege levels and 853 agent tasks, finding current models perform poorly at 40% accuracy. This highlights the need for scalable conflict resolution in complex agentic environments.
</span>
</span>
</div>
</td></tr></tbody></table>
</td></tr></tbody></table>
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding-top: 0px; padding-bottom: 0px;">
<div class="text-block">
<div style="text-align: center;"><span style="font-size: 36px;">🎁</span></div></div>
</td></tr></tbody></table>
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding-top: 0px; padding-bottom: 0px;">
<div class="text-block">
<div style="text-align: center;"><strong><h1>Miscellaneous</h1></strong></div>
</div>
</td></tr></tbody></table>
<table bgcolor="" style="table-layout: fixed; width: 100%;" width="100%"><tbody><tr><td style="padding:0;border-collapse:collapse;border-spacing:0;margin:0;" valign="top">
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fwww.dwarkesh.com%2Fp%2Fjensen-huang%3Futm_source=tldrai/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/RFu7jGRIOQ0gexLNijRHk1NuIiecx1azDE3hDhTU7io=452">
<span>
<strong>Jensen Huang – TPU competition, why we should sell chips to China, & Nvidia's supply chain moat (90 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
This post features a transcript of an interview with Jensen Huang. He discusses TPU competition, Nvidia's lock on the supply chain needed to make advanced chips, whether the US should sell AI chips to China, why Nvidia isn't a hyperscaler, how the company makes its investments, and more. Links to audio and video of the interview are available.
</span>
</span>
</div>
</td></tr></tbody></table>
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fwww.implicator.ai%2Fclaude-probably-wasnt-secretly-nerfed-anthropic-made-the-black-box-too-dark%2F%3Futm_source=tldrai/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/XGLv1wlWf6I28ab6RaxSKwU32m98o9v9zFR4Z58iWjE=452">
<span>
<strong>Claude probably wasn't secretly nerfed. Anthropic made the black box too dark (10 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Users have accused Anthropic of nerfing Claude Code, but there's no evidence that Anthropic has done this. The strongest public reports still lack independent raw data. However, Anthropic didn't need to nerf Claude for Claude Code to become a different product. Effort defaults, adaptive thinking, cache duration, context compaction, quota policy, and status incidents can all change the experience while the model name stays the same.
</span>
</span>
</div>
</td></tr></tbody></table>
</td></tr></tbody></table>
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding-top: 0px; padding-bottom: 0px;">
<div class="text-block">
<div style="text-align: center;"><span style="font-size: 36px;">⚡</span></div></div>
</td></tr></tbody></table>
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding-top: 0px; padding-bottom: 0px;">
<div class="text-block">
<div style="text-align: center;">
<h1><strong>Quick Links</strong></h1>
</div>
</div>
</td></tr></tbody></table>
<table bgcolor="" style="table-layout: fixed; width: 100%;" width="100%"><tbody><tr><td style="padding:0;border-collapse:collapse;border-spacing:0;margin:0;" valign="top">
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fjobs.ashbyhq.com%2Ftldr.tech%2F3b21aaf8-dea5-4127-be71-602d30e5001e%3Futm_source=tldrai/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/5jg-2JjXToeWvSQ8p9LIuMAncNaa5jB4hWtU6Dd-dLE=452">
<span>
<strong>TLDR is hiring a Senior Software Engineer, Applied AI ($250k-$350k, Fully Remote)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
TLDR's Applied AI team is tasked with making every process at TLDR legible to code, runnable by anyone, and composable into larger workflows. Join a small, fast moving team using the latest AI tools with an unlimited token budget. <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fjobs.ashbyhq.com%2Ftldr.tech%2F3b21aaf8-dea5-4127-be71-602d30e5001e/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/Aqiohe2it9HsBMuzkrQduaR4M9MtQHBWQNTTkjpJRXo=452" rel="noopener noreferrer" target="_blank"><span>Learn more</span></a>.
</span>
</span>
</div>
</td></tr></tbody></table>
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fblog.cloudflare.com%2Fbrowser-run-for-ai-agents%2F%3Futm_source=tldrai/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/7FMnRMg4XoFjcb7BGZ3Gse58GOScZRqfm3z5Gvwvirk=452">
<span>
<strong>Cloudflare Browser Run: Edge-Native Infrastructure for AI Agents (12 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Cloudflare has rebranded its Browser Rendering service to BROWSER RUN, introducing new features for AI agents.
</span>
</span>
</div>
</td></tr></tbody></table>
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fwww.testingcatalog.com%2Fgoogle-tests-agentic-shopping-with-native-checkout-in-gemini%2F%3Futm_source=tldrai/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/oZpU9sWdN5aMFt6luUtzDRZpzkb1mGDf4qSIaQRZfx0=452">
<span>
<strong>Google tests Agentic Shopping and native checkout in Gemini (2 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Google is testing a Shopping Cart feature within the Gemini app, allowing in-app product browsing and purchasing.
</span>
</span>
</div>
</td></tr></tbody></table>
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fblog.google%2Finnovation-and-ai%2Fproducts%2Fgemini-app%2Fgemini-app-now-on-mac-os%2F%3Futm_source=tldrai/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/Xa0qxiREuY9o2_2eHMtjWbLnb7DmxuC59eTTeXgFKh4=452">
<span>
<strong>Gemini App for macOS (3 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Google released a native Gemini app for Mac with system-wide access, screen context sharing, and support for image and video generation using Nano Banana and Veo.
</span>
</span>
</div>
</td></tr></tbody></table>
<table align="center" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block">
<span>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fthenextweb.com%2Fnews%2Fjane-street-coreweave-6-billion-cloud-1-billion-equity-ai%3Futm_source=tldrai/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/0QiMj4GDLPDXJOdnx6Z9yuv3lBLRj1cyEwSREQimYY4=452">
<span>
<strong>Jane Street commits $6 billion to CoreWeave and takes a $1 billion equity stake (7 minute read)</strong>
</span>
</a>
<br>
<br>
<span style="font-family: "Helvetica Neue", Helvetica, Arial, Verdana, sans-serif;">
Jane Street is betting that the returns on AI-driven trading will justify $6 billion in cloud spending and a $1 billion equity position in the company providing it.
</span>
</span>
</div>
</td></tr></tbody></table>
</td></tr></tbody></table>
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td align="left" style="word-break: break-word; vertical-align: top; padding: 5px 10px;">
<p style="padding: 0; margin: 0; font-size: 22px; color: #000000; line-height: 1.6; font-weight: bold;">
Love TLDR? Tell your friends and get rewards!
</p>
</td></tr>
<tr><td class="container" style="padding: 0px 10px 15px;">
<div class="text-block">
Share your referral link below with friends to get free TLDR swag!
</div>
</td></tr>
<tr><td align="left" style="padding: 10px;">
<div class="text-block">
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Frefer.tldr.tech%2F51d793d6%2F2/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/JWdPpv8sYWtbm85QoD2ma0j9r656HeF07O81R8YVCe4=452" style="color: #464ba4; text-decoration: underline;">https://refer.tldr.tech/51d793d6/2</a>
</div>
</td></tr>
<tr></tr>
<tr><td align="left" style="padding:5px 10px;">
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fhub.sparklp.co%2Fsub_9a4c16be99eb%2F2/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/bxgYzgRmy6AIaBZGbQ4IH26ZpVEXzsmI3PwtA0XPVbA=452" style="font-size: 16px; line-height: 1.6; padding: 10px 0; display: inline-block; text-decoration: underline;"><span style="mso-text-raise:13pt; text-decoration: underline;">Track your referrals here.</span></a>
</td></tr></tbody></table>
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td align="left" style="word-break: break-word; vertical-align: top; padding: 5px 10px;">
<p style="padding: 0; margin: 0; font-size: 22px; color: #000000; line-height: 1.6; font-weight: bold;">
Want to advertise in TLDR? 📰
</p>
<div class="text-block" style="margin-top: 10px;">
If your company is interested in reaching an audience of AI professionals and decision makers, you may want to <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fadvertise.tldr.tech%2F%3Futm_source=tldrai%26utm_medium=newsletter%26utm_campaign=advertisecta/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/hgeaEOJEGvFrQHXRZglqnFNNSSur0dZeizkwidueJug=452"><strong><span>advertise with us</span></strong></a>.
</div>
<br>
<!-- New "Want to work at TLDR?" section -->
<p style="padding: 0; margin: 0; font-size: 22px; color: #000000; line-height: 1.6; font-weight: bold;">
Want to work at TLDR? 💼
</p>
<div class="text-block" style="margin-top: 10px;">
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fjobs.ashbyhq.com%2Ftldr.tech/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/DqSge8bqp0XPWopnErO764pENYluJZOz_rzUZ6loaiA=452" rel="noopener noreferrer" style="color: #0000EE; text-decoration: underline;" target="_blank"><strong>Apply here</strong></a>,
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fjobs.ashbyhq.com%2Ftldr.tech%2Fc227b917-a6a4-40ce-8950-d3e165357871/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/YFh9daJd1eAlKqF0fjSYpltodxZGj81p3GU7zLseqbY=452" rel="noopener noreferrer" style="color: #0000EE; text-decoration: underline;" target="_blank"><strong>create your own role</strong></a> or send a friend's resume to <a href="mailto:jobs@tldr.tech" style="color: #0000EE; text-decoration: underline;">jobs@tldr.tech</a> and get $1k if we hire them! TLDR is one of <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fwww.linkedin.com%2Ffeed%2Fupdate%2Furn:li:activity:7401699691039830016%2F/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/Q9ZDOmz3VqwQLoWFoX3JNKozZm--37ZBBjF7Qb_NtJc=452" rel="noopener noreferrer" style="color: #0000EE; text-decoration: underline;" target="_blank"><strong>Inc.'s Best Bootstrapped businesses</strong></a> of 2025.
</div>
<br>
<div class="text-block">
If you have any comments or feedback, just respond to this email!
<br>
<br> Thanks for reading,
<br>
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Ftwitter.com%2Fandrewztan/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/rTiGLLMwmp-pSm0Rq6jpBHWF8eB07I2NPvUbmYsdpoY=452"><span>Andrew Tan</span></a>, <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fwww.linkedin.com%2Fin%2Faliiaminian%2F/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/jvwBuwHIBXrrsrsKGZSf3FpPryyrnQAG906g-jc5XgU=452"><span>Ali Aminian</span></a>, & <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fwww.linkedin.com%2Fin%2Fjacob-turner-7521a8198%2F/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/yrQrdkWx041qP98I3fVKJLoIprkYw_MMcfW5PBRvV-A=452"><span>Jacob Turner</span></a>
<br>
<br>
</div>
<br>
</td></tr></tbody></table>
<table align="center" bgcolor="" border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td class="container" style="padding: 15px 15px;">
<div class="text-block" id="testing-id">
<a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Ftldr.tech%2Fai%2Fmanage%3Femail=silk.theater.56%2540fwdnl.com/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/-jitdMKHoYffyninDS41snVtKC1QIg3hL3I3MjszLx8=452">Manage your subscriptions</a> to our other newsletters on tech, startups, and programming. Or if TLDR AI isn't for you, please <a href="https://tracking.tldrnewsletter.com/CL0/https:%2F%2Fa.tldrnewsletter.com%2Funsubscribe%3Fep=1%26l=eedf6b14-3de3-11ed-9a32-0241b9615763%26lc=1571a9f6-84b7-11f0-b464-47c5c04ad337%26p=b97b2e02-3978-11f1-85a8-e709599e4e50%26pt=campaign%26pv=4%26spa=1776344636%26t=1776345636%26s=0872aff117e1fbfe432de0bf40c912179af3b7b9f97132a6d7f26292d8d13479/1/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/5z3WrPBoBNK8aHtmi3IiRGXLG9YKfrU6PfZoeN2u5QM=452">unsubscribe</a>.
<br>
</div>
</td></tr></tbody></table>
</td></tr></tbody></table>
</td></tr></tbody></table>
</td></tr></tbody></table>
</td></tr></tbody></table>
<img alt="" src="http://tracking.tldrnewsletter.com/CI0/0100019d96735e2a-ed569f0f-9093-4977-84fb-5dc151cc1fe4-000000/dLBTe9jTbvhiJ2Gd9g8KffFMyYHGoKfOOIk9mKdgPvs=452" style="display: none; width: 1px; height: 1px;">
</body></html>