Nebius Group: KI Infrastruktur und KI Dienstleistungen
Beliebteste Beiträge
Beitrag zu dieser Diskussion schreiben
Nutzer sind auch aktiv in:
- 1
- ►
- …
- 30
Begriffe und/oder Benutzer
Top-Postings
https://knowyourinvestments.substack.com/p/nebius-watch-brea…
"Ende April veröffentlichte Nebius seinen Jahresbericht 2025 (Formular 20-F), der einige interessante, bisher unveröffentlichte Details enthielt. Ich habe das Formular 20-F analysiert und werde in diesem Artikel die wichtigsten Erkenntnisse hervorheben, insbesondere:
- Die 40%ige Vorauszahlungsrate gemäß dem Microsoft-Vertrag sowie meine Gedanken zu Vorauszahlungsstrukturen in Meta-Verträgen.
- RPO-Zahlen , einschließlich der Tatsache, dass rund 1 Milliarde US-Dollar des RPO zum Jahresende 2025 von anderen Kunden als Microsoft und Meta stammten.
- Die genaue Marktkapitalisierung von Nebius basiert auf der vollständig verwässerten Aktienanzahl (Spoiler: Sie beträgt nicht 45 Milliarden Dollar, wie viele Finanzdatendienste suggerieren).
- Mögliche Risiken im Zusammenhang mit der Frage, ob Arkady Volozh aufgrund einer Verwässerung die Kontrolle über das Unternehmen verlieren könnte.
- Geschäftliche Neuigkeiten zu TripleTen und Avride sowie Beteiligungen an ClickHouse und Toloka.
- Abschließend noch ein Update zu meiner fairen Bewertung der Nebius Group."
10 % hoch und das ohne große Neuigkeit?
"Scaling AI at Inference: The Road to Agent-Driven ROI"
00:00:00 - Introduction to AI and Infrastructure
00:00:28 - The Evolution of AI and Its Impact
00:01:33 - Introducing Roman from Nebius
00:02:29 - Nebius: AI-Specialized Cloud
00:03:45 - Shifting Focus from Training to Inference
00:05:25 - Customer Behavior and Infrastructure Needs
00:10:12 - Challenges in Closed Ecosystems
00:11:39 - The Role of Open Source in AI
00:17:28 - Making Open Source Enterprise-Grade
00:22:04 - Differentiating Nebbius in the Market
00:23:02 - Customer Engagement and Support
00:29:02 - Use Cases for AI Agents
Ein Nebius Mitarbeiter auf X:
I work on inference for a living, at
@nebiustf
, where we run open-source managed inference at scale. Most of what follows is what I'm seeing from inside the bill.
12 months ago, the cost of 1M tokens of frontier-class reasoning was somewhere on the order of $60.
Today, an equivalent quality of output costs roughly $0.50.
Price /token of o1-level intelligence has dropped about a 128x in a year.
Price of GPT-4-level output has dropped roughly 100x since the original GPT-4 shipped.
By any normal reading of a technology cost curve, this should be deflationary. It should be saving customers money.
The opposite has happened. The total compute bill at every hyperscaler is going up, not down. Anthropic just signed multi-year capacity deals with both XAI and Amazon. Microsoft's Azure capex guide for 2026 starts with an eight. OpenAI is reportedly spending more on compute every quarter than it did in all of 2023. Nvidia paid roughly twenty billion dollars to acquire Groq, an inference-specialist company that did not exist as a serious commercial entity three years ago.
The cost curve and the demand curve crossed, and then the demand curve lapped the cost curve.
Here is what happened underneath.
A reasoning model burns roughly 10x the output tokens of a non-reasoning model on the same task, because it spends most of its tokens thinking out loud before answering. An agentic workflow chains roughly twenty times the requests of a single-shot completion, because it loops, calls tools, plans, retries, and synthesizes. A modern deep-research query (the kind a research analyst can fire off in fifteen seconds and then walk away from for ten minutes) costs more compute than 10 original GPT-4 queries combined. We made every individual token a hundred times cheaper, and then we built a generation of products that consume ten thousand times more tokens.
This is the Jevons paradox playing out at trillion-dollar scale, in compressed time, in front of everyone. Jevons noticed in 1865 that making coal-burning more efficient did not reduce coal consumption. It increased it, because efficiency unlocked uses that were previously uneconomic. Steam engines became more practical at smaller scales. Whole industries that could not afford coal at the old price suddenly could. Britain's coal consumption rose sharply, not despite the efficiency gains, but because of them.
The same thing is happening to AI compute right now and it is happening faster than any analogous historical cycle. Falling token prices did not contract demand. They unlocked agents, deep research, code-writing systems, multi-step reasoning, persistent memory, the entire next layer of AI products. Every product in that next layer consumes orders of magnitude more compute than the chat interfaces it is replacing.
The math at the aggregate level is brutal: 100x cheaper tokens times 10 000 more tokens equals a 100x larger total bill.
The implications stack quickly.
If you are running a hyperscaler, your 2026 capex guide is not a peak. It is a step on a curve. Inference is structurally always-on, twenty-four hours a day, in a way that training never was. Training is bursty. You spin up a cluster, run for weeks or months, and stop. Inference runs continuously, scales with usage, and the usage curve is exponential. Your power bill, your cooling bill, your transceiver count, your storage footprint, all of these were sized for a workload mix that no longer exists.
If you are running an AI software company built on top of someone else's closed API, you have a problem that did not exist a year ago. Your gross margins get worse as your customers get more value out of your product, because the more they use it, the more compute you pay for. The companies that win this are the ones that figured out vertical integration before the math caught them.
If you are watching this from a distance and trying to understand where the next bottlenecks form, the answer is everywhere downstream of "more inference compute, always-on, with massive memory state per session." The KV cache, the running memory state of a long conversation or an agent loop, is the silent monster of the inference era. It does not scale linearly with parameters. It scales linearly with context length and number of agent steps. A long agent session can hold tens of gigabytes of state per user, per session.
Multiply that by every concurrent user of every product, and you understand why
$MU
,
$SNDK
, $TOWCF, and the entire memory and packaging layer have re-rated the way they have.
The CPU-to-GPU ratio is evolving. Training is 1:8. Basic chat inference is 1:4. Agentic inference is 1:1, sometimes CPU-heavy. Google has split its TPU line in two, with a dedicated inference chip carrying tripled SRAM for KV cache.
$INTC
and
$AMD
just spent two earnings calls explaining that this shift is structural, not cyclical. The hardware map is redrawing in real time and the financial press is mostly still writing about training clusters.
The right framing of where we are right now is not that AI is hitting a wall. The framing a year ago that scaling was hitting a wall was the most expensive bad take of the cycle. The right framing is that AI got dramatically cheaper, dramatically more capable, and dramatically more useful, and the cost of running it at the new equilibrium of demand is much higher than the cost at the old equilibrium of demand, because the new equilibrium is enormous.
A meaningful share of what we actually do at Token Factory, day to day, is help customers stop their bills from running away from them. KV-cache management. Speculative decoding. Quantization. Routing. The kind of vertical integration that, eighteen months ago, every product team was happy to leave abstracted away behind a closed API. The reason this stack matters now is the same reason this whole essay matters: at the new equilibrium of inference demand, the cost of treating compute as a commodity is no longer survivable. The companies that figure out the layer beneath the API are the ones who keep their margins.
Cheaper tokens. More tokens.
Same coal as 1865."
Danke!
"Schlussbetrachtung
Alles in allem bin ich wohl so zuversichtlich wie noch nie vor der Veröffentlichung eines Nebius-Quartalsberichts. Nicht im Hinblick auf die Kursentwicklung, die ich nicht vorherzusagen versuche, sondern im Hinblick auf die tatsächliche Unternehmensleistung.
In den vergangenen Quartalen befürchtete ich, dass Analysten den Kapazitätsausbau zu früh eingeschätzt hatten. Diesmal glaube ich, dass das Gegenteil der Fall sein könnte. Nebius beendete das Jahr 2025 mit mehr aktiver Kapazität als erwartet, lieferte im Februar zusätzliche Kapazitäten für Microsoft aus, sicherte sich weiteres Kapital und operiert weiterhin in einem der stärksten Nachfrageumfelder, die ich je erlebt habe.
Die Aktie hat bereits einen enormen Kursanstieg hingelegt, daher ist mit Volatilität zu rechnen, und ein positives Ergebnis dürfte kurzfristig bereits eingepreist sein. Ich sage nicht, dass die Aktie nach der Veröffentlichung der Quartalszahlen sprunghaft ansteigen wird. Aber fundamental gesehen geht Nebius meiner Meinung nach aus einer starken Position in die Berichtsveröffentlichung.
Wenn das Management eine kontinuierliche Umsetzung nachweisen, Fortschritte bei der Erreichung seiner Kapazitätsziele für 2026 und 2027 bestätigen und demonstrieren kann, dass die Nachfrage weiterhin deutlich über dem verfügbaren Angebot liegt, glaube ich, dass sich die These weiter stärken wird."
AkibaBC
😂🤣😂🤣😂🤣😂🤣
Gratuliere! Du bist der Allergrößte!
Wer meine Posts bei dieser Aktie im letzten Sommer verfolgt, die Aktie gekauft und bis dato gehalten hat, dürfte sich nunmehr mächtig freuen!💶
Ich habe vor ca. zwei bis drei Wochen das gleiche bei Seagate und Samsung gemacht. Ich ärgere mich noch immer schwarz deswegen.
Bin ein Fan, aber habe dennoch heute einige Gewinne mitgenommen heute, die Bewertung ist mit 38 Milliarden verdammt hoch für ein Unternehmen, welches kaum schwarze Zahlen schreibt, hoch verschuldet ist und die US-Wirtschaft könnte sich im 2. Quartal auch abkühlen.
https://discord.com/channels/1337857173182611519/13382482372…
Bei den Vorauszahlungen sollte sich doch eine KE vermeiden lassen.
Ich denke, dass geht echt auf die Meldung über die Vorauszahlung von Microsoft zurück.
Krassomat heute.
https://x.com/zen_tropy/status/2050945242384277876/photo/1
11.05.26 · Sharedeals · Nebius Group Registered (A) |
11.05.26 · wallstreetONLINE Redaktion · Applied Materials |
19.04.26 · Directors Dealings · Sixt Vz |
16.04.26 · Markt Bote · Nebius Group Registered (A) |
