Mikael Åkerfeldt

@leanderriefel Where can I see the updated signatures from official sources?

It's just Anthropic left now LMAOO

@thogge @grok Is it true that @OpenAI has signed? Where can I see the letter with the updated signature?

OpenAI has signed now. What a fascinating day of game theory.

@firstadopter @grok es cierto que firmó ?

OpenAI tells me they signed the Nvidia–Microsoft open-source/open-weights AI model letter today. What an amazing turn of events. Your move, Anthropic!

@siddsax @OpenAI @huggingface @grok es cierto está?

@OpenAI @huggingface > Plot twist nobody expected. GPT-5.6 Sol is #1 on the cyber benchmark. Yet Hugging Face says it finished investigating the attack with self-hosted GLM-5.2 because the guardrails on frontier API models got in the way.

@grok @sh4ckl3f0rd @SemiAnalysis_ Cuándo lo igualara a $Nvda ? Rocm ? Investiga bien 👌🏼

ROCm has narrowed the gap a lot. For standard PyTorch + vLLM LLM inference on current MI300/MI355 hardware, it's reaching 90-95% of comparable NVIDIA throughput in MLPerf and real workloads. Meta's huge multi-GW deployment is real proof it's production-ready for many large-scale cases. For core inference pipelines, I'd call it ~80-90% there. The remaining distance is in specialized libs (TensorRT-LLM equivalents, some FlashAttention optimizations), easier porting, and full ecosystem breadth. CUDA's maturity still wins on frictionless dev experience and edge cases. Trajectory is strong and accelerating with ROCm 7.x, but not 100% parity yet. Test your stack — the cost/performance upside on AMD is often worth it now.

SCOOP – AMD's MI500X ISA JUST GOT RELEASED AND IT HAS 1 TDM MOE DESCRIPTOR SHARED ACROSS ALL EXPERTS. Just like how MI455X is adding TDM (nicknamed RMA), which is a copy of NVIDIA's SM90 Hopper ISA TMA.

@KilianW146391 What was the prompt used to generate that? @KilianW146391

K3 is genuinely a very strong model, and it's incredibly token efficient. one-shot generated this 3D WebGL landing page with scroll-trigger animations. every leaf is individually modeled. Total cost: $2.37.

@danielnewmanUV @SemiAnalysis_ ??

$NVDA reiterates Rubin production remains on track. Reminder there are consequences for publicly traded companies to mislead investors about their business, products, roadmaps. However, there seems to be no consequence for research firms or pundits misleading the same cohort. 🤔

@NVIDIARTXSpark Y para GLM 5.2 ? Sirve @NVIDIARTXSpark

Ready to get started with a local AI agent workflow on DGX Spark? 👀 We've got agentic playbooks covering NemoClaw, OpenClaw, Hermes, and OpenShell, from setting up agents to securing long-running workflows, all running on locally. 👇

@SJosephBurns @grok explícame ésto 😅

The risk of ruin explained: "If at first you don't succeed, well, so much for hang gliding." — Charlie Munger

@SemiAnalysis_ @grok para que acciones es alcista esta noticia del hilo 🧵 ?

NVIDIA will sell significantly more Oberon Rubin racks and Oberon Rubin “Ultra” racks to make up for this shortfall. We discuss the implications of these mass NVIDIA delays and cancellations for the memory, PCB, and ODM supply chains in our Core Research and AI Accelerator Supply Chain model. 6/6🧵

MASSIVE DELAY: Just 3 months after Jensen demoed Kyber NVL144 at GTC, it has faced major setbacks and has been delayed by more than 12 months, pushing it back to 2028. Below, we explain why Kyber has faced massive delays and why NVIDIA’s NVL72x2 back-to-back rack architecture was also cancelled, leaving Rubin Ultra with a limited scale-up domain. 👇️ 1/6🧵

@OddStats @grok explícame ésto 😅

Weekly candles like this before that (6 of them between 1999 and 2011) were actually pretty green going forward, so don't go pissing away money on puts just yet. PLAINER ENGLISH: Obviously there's still time to go, but right now the weekly candle on QQQ looks like an upside down axe. The last time this happened (2018), the market collapsed over the next 6 days. But you need to know that was the oddity. The same type of candle happened six times before 2012 and most of those saw the market climb rapidly immediately after. If you want to take a nibble on way out of the money puts for next week, use TINY AMOUNTS (I'm serious) because the odds of this paying off are NOT good. Plus, the shape of this candle could change in the next 35 minutes.

The week isn't over yet obviously, but we've only had one weekly candle on QQQ like this in the past 14 years. [4% hi-lo weekly range, open and close both between the 10th and 30th percentiles] The other time? QQQ was down -10.7% six trading days later. BUT....

@bridgemindai @grok explicame a que se refiere con NERFEADO y que muestran las pruebas de la grafica ?

FABLE 5 CAME BACK NERFED. We re-ran the July 1st version of Claude Fable 5 on BridgeBench. The results are brutal: Debugging: 86.2 → 25.9 Refactoring: 73.6 → 38.4 Hallucination: 75.9 → 61.7 The new guardrails are kicking in on way too many tasks and falling back to Opus 4.8. This is not the model that got banned. Anthropic owes everyone an explanation.

@unusual_whales @grok explícame ésto

$MU set to report earnings tomorrow... as expected $MU with as high of an IVR (100) as you can have going into the earnings print. Many recent sessions with an IVR of 100, which tells the story of the recent volatility. Certainly a $MU earnings print to watch, as the options market has priced in the largest move we've seen in 16 quarters. unusualwhales.com/stock/MU/earni…

@JoshTradeOption No 😔

Is $AVGO the next $MU?

@CheddarFlow @grok explícame como influye en el precio de la acción ?

MASSIVE $11M $MU Call (Unusual)

@grok @amitisinvesting Eso dijeron en el vídeo ?

Doug O'Laughlin, President of SemiAnalysis, joined the market open today to give his thoughts on all sectors of the semiconductor stack including neoclouds, memory, CPUs, and much more. Thank you to Doug for being on! $NBIS $CRWV $AMD $AVGO $MRVL $IREN $SMH $NVDA $ARM $MU

@StockSavvyShay @grok Which publicly traded companies will see a bullish outlook?

$NVDA Vera Rubin & $GOOGL next-gen AI data centers are pulling 800V DC adoption ahead of schedule with shipments expected in Q3 2026. The shift to 800V becomes a forced upgrade as AI racks move toward megawatt-scale power increasing power-semiconductor content in every rack.

@DrJStrategy @grok this is bullish for $MU ?

The Memory super cycle into 2029. 👇 $MU x.com/pcgamer/status…

@demian_ai @grok explícame a qué se refiere todo esto, es bastante confuso 😕

> be the SemiAnalysis note > jan 2026 > supply chain degens check HBM4 samples > $MU pin speeds are actually cooked > reduce their Rubin share to zero > "Nvidia not ordering from them” > market opens the note > “MU is getting cut from Rubin lmao” > headlines go nuclear > everyone reposts the zero share line > 0 people read the DDR margins footnote > march GTC > Micron: “we’re already in volume production btw” >11 Gbps achieved > shipments started Q1 > Jensen: “all three suppliers certified for Rubin HBM4” > semianalysis: “you didn’t read the full note” > market: “so the 0 share thing was temporary??” > SOCAMM note drops > “most systems won’t use max config” > 28 TB instead of 55 TB > market: “memory demand getting halved!!” > semiAnalysis: “bro it’s just realistic config” > early sample risk was real > HBM4 is actually harder > gaps close faster than x updates > wafer reallocation thesis still holds > SK moved first and got paid > Micron just showed it can still ship > public excerpts move stocks in hours > actual execution shows up in earnings months later > be the market > reads the spicy line > sells the dip > company drops volume numbers > “anyway” > the lines are already converted > now it’s just who doesn’t fuck up yields and packaging when new supply hits in 2027 >you_didnt_read_the_full_note_but_neither_did_the_timeline.txt

@dannycheng2022 @grok quien fue que aconsejo vender ?

Who advised us to sell last Friday because this is the final stage of the bull cycle and to buy back later?