Lambda in talks to raise up to $3B at a $12B-plus valuation

The neocloud fundraising wave keeps rolling. Nvidia-backed Lambda is reportedly in talks to raise up to $3 billion in a pre-IPO round that would value the GPU cloud provider above $12 billion — with revenue this year already estimated to top $1.5 billion, per Bloomberg. WSJ separately reports the company has raised over $1.5 billion, putting one of the largest independent AI compute players firmly on a path to a public listing.

Reuters, Bloomberg and WSJ report that Nvidia-backed neocloud Lambda is closing in on a pre-IPO raise of up to $3 billion at a valuation above $12 billion, with revenue this year estimated to exceed $1.5 billion. Lambda runs one of the dominant independent clouds for renting out Nvidia GPUs to AI startups and research labs, and the financing marks the latest sign that the capital markets are wide open for the infrastructure layer beneath the AI build-out. A valuation north of $12 billion would roughly represent an eight-fold-plus jump on its revenue run rate, a rich multiple for a business still scaling capacity — but one that reflects how scarce reliably provisioned GPU supply has become.

The raise matters beyond Lambda itself. It lands alongside a drumbeat of comparable moves across the so-called neocloud sector — CoreWeave and Nebius already trade publicly, FluidStack recently drew a $1 billion round at an $18 billion valuation, and even Groq, freshly pivoted from chips toward cloud, just pulled in its own financing. Every one of these players is betting the same thesis: that hyperscaler-native cloud isn't enough, and that a fast-growing tier of specialists can profitably rent out the newest silicon to whoever needs it. Lambda's pre-IPO push suggests the sector's founders believe those valuations are real enough to underwrite an exit.

For readers who track GPU supply and AI pricing, the number to watch is the revenue trajectory: $1.5 billion estimated this year, on prior plans to fund hardware through creative debt structures. Lambda has been aggressively expanding capacity — we covered its $917 million loan to finance an Nvidia GPU fleet — and this equity round is the natural follow-on to that credit-driven growth. If the deal closes anywhere near the reported terms, it gives Lambda the war chest to keep ordering Nvidia's newest chips at a time when every breakthrough model release tightens the market for compute.

With hosting costs now the single biggest variable in AI's scaling story, do you think current neocloud valuations are a sustainable bet or a bubble waiting for the first big correction? Tell us in the comments.

AI 101 — How do AI video generators work?

An AI video generator is a model that starts with a grid of random noise and repeatedly refines it into a moving picture that matches your prompt — the same trick image generators use, stretched across time. Nothing is filmed, and nothing is stitched together from clips the model has seen. The pixels in the finished video were invented, frame by frame, all at once. Why it matters right now Kling 4.0, Kuaishou's flagship video model, went fully live on October 15 with up to 30 seconds of vid

Houmo moves compute inside memory — and A20 enters production

Beijing chipmaker Houmo used its partner conference this week to lay out two compute-in-memory routes at once — and to say the first chip built on the newer one is already in production. Houmo's A20, its first chip built on a 3D compute-in-memory architecture, has entered production. The part supports up to 16 chips fully interconnected, claims a 50% improvement in compute utilization over the previous generation, and covers models from 3B to 30B parameters, according to co-founder and product

Kling 4.0 goes live with 30-second native video at 10-bit HDR

Kuaishou's Kling AI has spent two years becoming the video model people actually finish projects with. Its new flagship, Kling 4.0, went fully live on October 15 — and this round is less about prettier frames than about the two things that make a generated shot usable in a real edit: length and control. Kling 4.0 generates up to 30 seconds in a single native pass, outputs 10-bit HDR at 4K, 1080p and 21:9, and takes up to 15 reference assets in one task. That budget splits across ten images, fi

Brain organoids classified Braille at 83% — in a trio, not alone

Two stories today: human brain tissue on a chip decoded a tactile signal, and the data-centre company behind some of the largest AI contracts in Europe turns out to be six months behind on its flagship site. Human forebrain organoids grown on a microelectrode array classified Braille letters with 61 percent accuracy on their own and 83 percent when the responses of three organoids were combined, in a peer-reviewed study from the University of Bristol. The team took event-based tactile readings

Read more

AI 101 — How do AI video generators work?

Houmo moves compute inside memory — and A20 enters production

Kling 4.0 goes live with 30-second native video at 10-bit HDR

Brain organoids classified Braille at 83% — in a trio, not alone