Modal and Baseten in talks at $15B and $26B, months after last rounds

Share
Modal and Baseten in talks at $15B and $26B, months after last rounds

Two inference platforms are asking investors to reprice the layer that sits between the model labs and everyone using them. The second story is Qualcomm's bid to put an always-on agent in your ears.


Modal Labs is in talks to raise at a roughly $15 billion valuation, and Baseten is negotiating a round that could value it at $26 billion. Modal raised $355 million in May at a valuation of $4.65 billion, so the new number is roughly triple it — four months later. Baseten's last round, in June, put it at $13 billion, which makes the new figure a clean doubling in three months. Bloomberg reported both talks on Wednesday, citing people familiar with the matter; Axios Pro had the Baseten half first, on Tuesday. Neither company has commented, and neither round has closed.

Why it matters: both companies sell the same scarce thing — the ability to run models and agents without owning a data center. Modal rents GPU capacity and provides sandboxes for agent code; Baseten serves models in production. Tripling in four months is not a product signal so much as a pricing signal about what it now costs to serve inference at scale. We looked at the other end of that trade earlier this month — Deep Dive — Tokens halved in price. The bill to make them didn't — and the two stories sit on the same ledger. If the price per token keeps falling while the platforms that deliver tokens double in value, the squeeze lands on whoever owns the hardware.


Qualcomm announced the Snapdragon Sound Elite Gen 2, an audio-wearable platform it says delivers up to twice the on-device AI performance of the previous generation, at up to 40% lower power draw and in a 30% smaller footprint. The chip adds Wi-Fi 6E and a dual Bluetooth radio, and Qualcomm brought Bose, HP and Cleer on stage as launch partners at Snapdragon Summit. Every performance number is the vendor's own "up to" claim: there are no shipping earbuds, no independent benchmarks, and the first devices are months away. Qualcomm is the same company that put a 30B-parameter model on a phone this month — Qualcomm's next NPU runs 30B-parameter models on a phone — and this is that strategy pointed at your ears.

The number worth watching is power, not AI. An always-on audio agent that listens, translates and answers in real time is a battery problem long before it is a model problem, which is why 40% matters more than 2x here. Earbuds are the least intrusive place to put an ambient agent and the hardest place to keep one alive.

What to watch: whether the first Snapdragon Sound Elite Gen 2 devices actually run their agents on the earbud, or quietly fall back to the phone in your pocket.

Would you pay extra for earbuds that run an agent locally, or is the phone you already carry enough? Tell us in the comments.

Read more

Akhetonics says its all-optical CPU reaches a customer in 2026

Akhetonics says its all-optical CPU reaches a customer in 2026

Light-based computing keeps promising more than it delivers — but one Munich startup has just put a date on its bet, and the interview laying it out is doing the rounds on Hacker News this week. Akhetonics says it will deploy its first commercial machine with a major customer by the end of 2026, with several more planned for 2027. The company, founded by Michael Kissner and Leonardo Del Bino, is building a computer where data enters as light, is switched as light, and circulates through memory

The Week in AI — October 5–11, 2026

The Week in AI — October 5–11, 2026

Every big claim this week turned out to rest on fine print more interesting than the headline: revenue only the company reporting it can define, safety tests sandboxed while the product keeps the web, and a Pentagon phase-out nobody would confirm until reporters kept asking. The week's top 5 1. OpenAI's revenue was $20 billion below the numbers everyone quoted — and the gap was definitional. The Financial Times reported Thursday that OpenAI's annualized revenue runs roughly $20 billion unde

Drone strike shuts a third Yandex data center, taking YandexGPT offline

Drone strike shuts a third Yandex data center, taking YandexGPT offline

Russia's largest tech company is learning what the AI era's infrastructure war looks like from the receiving end — three data centers in four days, and with them much of the cloud layer Russian businesses run on. A Ukrainian drone strike knocked out Yandex's data center in Vladimir early Sunday morning, the third of the company's facilities hit since October 8. The site — reported at roughly 50 MW and designed for about 2,880 server racks — stopped operating completely after the attack, Yandex

Agent teams cost up to 5x more, barely score higher

Agent teams cost up to 5x more, barely score higher

The multi-agent hype train hit a benchmark this weekend — and the grid and the trucking regulators had quiet weeks of their own. Vals AI put agent teams head-to-head with single agents on its Vibe Code Bench, and the teams cost between 1.8 and 5.1 times more for almost no extra quality. The evals company ran GPT-6 Sol and Claude Opus 5.5 solo and in teams across 50 apps at two reasoning efforts; out of four comparisons, only one was statistically significant — Sol at medium effort, where the t