AI powered Equity Research platform

Joined October 2025
The market is underwriting the spend and waiting, or it’s tolerating it and repricing on the margin. Curious which one people are on. Meta Capex, Q2 2025: $16.5B Meta FCF, Q2 2025: $9B Meta Capex, Q2 2026: $30B Meta FCF, Q2 2026: $1.7B $META
20
8
1
30
28,370
Astreus retweeted
BREAKING: Elon Musk says $SPCX ARR could reach $500B-$1T
2
2
53
BREAKING: Elon Musk says $SPCX ARR could reach $500B-$1T
2
2
53
$NBIS publishes live GPU spot prices across it's cloud business! Can track the progress of data centre capacity + GPU spot prices for Nebius on Astreus!
1
1
2
118
$IREN Is probably the best positioned Neocloud on the sole basis that it has the power to turn on its data centre capacity. $CRWV is leading with roughly 850MW of electricity today but IREN has 2x the amount contracted in the future.
KEN GRIFFIN JUST DISCLOSED A 5% PASSIVE STAKE IN $IREN
8
1
29
2,125
No John Ternus $AAPL ? Also missing $AVGO $INTC $ORCL
Pretty big WH meeting on AI: Jeff Bezos (Chairman) - $AMZN Elon Musk (CEO) - $SPCX $TSLA Mark Zuckerberg (CEO) - $META Jensen Huang (CEO) - $NVDA Sundar Pichai (CEO) - $GOOGL Satya Nadella (CEO) - $MSFT Lisa Su (CEO) - $AMD Sanjay Mehrohtra (CEO) - $MU Alex Karp (CEO) - $PLTR Nikesh Arora (CEO) - $PANW Dario Amodei (CEO) - @AnthropicAI Greg Brokman (President) - @OpenAI
4
5
649
Pretty big WH meeting on AI: Jeff Bezos (Chairman) - $AMZN Elon Musk (CEO) - $SPCX $TSLA Mark Zuckerberg (CEO) - $META Jensen Huang (CEO) - $NVDA Sundar Pichai (CEO) - $GOOGL Satya Nadella (CEO) - $MSFT Lisa Su (CEO) - $AMD Sanjay Mehrohtra (CEO) - $MU Alex Karp (CEO) - $PLTR Nikesh Arora (CEO) - $PANW Dario Amodei (CEO) - @AnthropicAI Greg Brokman (President) - @OpenAI
1
1
3
932
BREAKING: Anthropic plans to spend $512B in AI-Infra (S-1) - REUTERS
2
191
BREAKING: Anthropic files S-1 (IPO Prospect) 2025 financials: Revenue : $4.59B Net (loss): $8.06B GAAP net loss: $41.97B, vs $8.31B in 2024 Anthropic spent $7.33B on compute and infrastructure in 2025, up 190% YoY and nearly 3x the prior year.
2
2
220
Introducing Transcript Search See key word mentions through 10,000+ earnings calls. Also search company tickers to see who’s talking about selected company. Live now at astreus.ai - 14 day free trial for all.
5
2
8
135
Monitor H100 spot rates across the Hyperscalers, Neoclouds, and the remaining marketplace on Prometheus today!
BREAKING: Some Nvidia customers have been notified that AI server prices will rise more than 15% Can now track GPU prices today on Prometheus!
5
1
12
261
Astreus retweeted
Replying to @AstreusAI
Using prometheus to track market data is like buying btc in 2015
2
2
49
1/2 Initial Compute futures registration
2
1
9
229
CrowdStrike $CRWD fiscal Q2’27 results: - Revenue: $1.47B vs. $1.44B expected, up 26% YoY - Adjusted EPS: $0.31 vs. $0.29 expected, up 35% YoY - Net new ARR: $333M, up 51% YoY - Free cash flow: $377.4M vs. $353M expected, up 33% YoY - ARR: $5.84B, up 25% YoY - Non-GAAP operating margin: 25% - Operating cash flow: $530.3M FY27 guidance: - Revenue: $5.99B–$6.01B vs. $5.93B expected - Adjusted EPS: $1.25–$1.26 vs. $1.23 expected - Operating income: $1.50B–$1.51B vs. $1.47B expected - Net income: $1.30B–$1.31B vs. $1.28B expected - ARR: $6.60B–$6.61B Q3 guidance: - Revenue: $1.52B–$1.53B vs. $1.51B expected - Adjusted EPS: $0.31 vs. $0.31 expected - Operating income: $372.7M–$375.9M vs. $369M expected - Net income: $325.4M–$327.9M vs. $321M expected - ARR: $6.18B–$6.19B Management raised its FY27 net new ARR growth outlook by 630 basis points, citing record net new ARR, record net new ARR from new customers, and improved dollar-based gross and net retention rates. Sources
1
2
8
197
NVIDIA $NVDA fiscal Q2 results: • Revenue: $96.2B vs. $92.2B expected • Adjusted EPS: $2.22 vs. $2.10 expected • Data center revenue: $89.0B, up 117% YoY; 93% of total revenue • Net income: $59.7B • EBIT: $63.7B • Gross margin: 75% Q3 guidance: $108B in revenue vs. $104.2B expected.
6
5
10
1,073
Measure this dynamic on Prometheus!
There’s a very interesting dynamic between the Neoclouds and the Hyperscalers Hyperscalers (especially Microsoft/Azure, and to varying degrees others) are significant customers of leading neoclouds such as CoreWeave, Nebius, and others. Microsoft has accounted for a very large share of CoreWeave’s revenue in recent periods (reports have cited figures around two-thirds in 2025) and has multi-billion to tens-of-billions in long-term capacity commitments across several neoclouds. Why this happens? Hyperscalers cannot stand up dense, high-performance GPU clusters fast enough to match AI demand. What does this mean for the markets? 1. Demand is outrunning even the biggest balance sheets 2. A new specialized layer has permanently inserted itself AI workloads (especially large-scale training clusters and increasingly inference) are different enough from general cloud that purpose-built providers can capture meaningful share and win anchor contracts from the hyperscalers themselves. The cloud market is no longer just the big three + Oracle; there is now a distinct GPU-native tier.
1
8
253