You saw how hard $HMM started to run on a small ETH test pump the other day… Wait until Ethereum rips back to $3K+ soon. The thinking cat will run to new all time highs with it. Bullieve in something 🤔🐈

Daily Memecoin Recap - August 28 Robinhood continues to dominate. Will CZ retaliate? Artificial Inu $ai -> hit $115m, nvidia paired memecoin on robinhood - absolutely insane run - random user "dumbcrayoneater" on fomo is up over $3,000,000 uPNL from a $21k buy Psyop Trump Coin $gold -> hit $62m - coin claiming to be affiliated with Trump. most likely fake. still managed to run to $60m+ - speculation that the domain expired today and someone bought it and launched a coin Robot Duck $microduck -> hit $9.3m, viral - Nvidia recently acquired Hugging Face - Hugging Face launched this small robot duck - Launched on robinhood, paired with nvidia Ape On Your Phone ATH $fone -> hit $38.7m, fomo x pump meme Breast Cancer Movement $pink -> hit $5.9m - fees donated to @BCRFcure - ppl on the pump app were changing their pfps to pink Robinhood $pons -> hit $111m $chump -> hit $24m $wallet -> hit $12.3m solana:GDAdCvjiRoGxA18ya3Qha1DD6bbzbR9Z9AGpMa9Tpump -> hit $8.6m $agi -> hit $5m $ntf -> hit $4m, career simulator $sit -> hit $3.8m, paired, paired with $ai $au -> hit $3.7m, paired with $tsm $ponsfolio -> hit $3.6m $vynex -> hit $3.5m $prometheus -> hit $1.6m $PDEX -> hit $1.6m solana:EghWat5y3EcCb7T4vxz276seEHSnrkEcH17Ey61Wpump -> hit $1.3m, “truth social index”, paired with $djt $gg -> hit $1m, golden goose $PLTR -> hit $940k More Plays $ansem -> hit $400m, steadily moving back to ath $toad -> hit $7.6m $bullshit -> hit $7.2m, ansem beta $baseline -> hit $5.2m, vaseline temporarily changed its brand to “Baseline". rugged $windchill -> hit $1.8m, cat/dog meme after a cat survived via dog blood transfusion $rent -> hit $1.5m $gta -> hit $1.5m $flork -> hit $1.5m $otc -> hit $1.4m $ast -> hit $940k, launchpad where each token is paired by a 2x–5x perp $apegun -> hit $915k another day where you could have made 6 figs

If you don’t agree with everything JD Vance says, you can probably expect @alexbruesewitz to run to the media about you to smear you with lies. I have been very supportive of the Trump administration and because I don’t agree with JD’s assessment of the Iran war, because I don’t believe in negotiating with terrorists, Alex says I am “viciously attacking the VP” and his allies. Alex will say he doesn’t work for JD Vance but he told me himself he works for 1789 Capital as an advisor to Omeed Malik. So sure, Alex can say he doesn’t work for JD, but he works for the people who will probably be funding JD’s Presidential campaign. I will keep this hit piece that Alex now pretends he had nothing to do with in mind for the future. @drewharwell told me on the phone that Alex was his “source” for his hit piece on me that was full of Hersey and hypocrisy since Alex is attacking me for work he asked me to do over a year ago. When I asked Alex if he planted a hit piece on me, his immediate reaction was accusing me of attacking JD, all because I don’t believe you can negotiate with Islamic terrorists and the MOU was always destined to fail. The negotiations failed. Alex can talk to as many reporters as he wants in an effort to smear me. It doesn’t erase the failure of the negotiations. It doesn’t erase the stench of a lot of things, including Alex.

Bret Johnsen — IT’S IN THE DNA at @SpaceX “There’s something about SpaceX. It’s kind of in our DNA. We love to run to those challenges. We brought back the first rocket booster from space. No one had done it before.”

Quantized Qwen3.8 27B for agentic coding: Same Score, 42 Different Outcomes.  I’ve been evaluating quantized Qwen3.8 27B models on DeepSWE 1.1, but I’m giving up. 4-bit AWQ run #1#: 38.94% Reward 4-bit AWQ run #2#: 31.86% Reward 32 of the 113 task outcomes flipped: 20 previously solved tasks failed, while 12 previously failed tasks passed. On non-agentic tasks, I’ve found NVIDIA NVFP4 performs very closely to AWQ. On DeepSWE 1.1? NVIDIA NVFP4: 31.86% Reward Exactly the same score as AWQ run #2#, but 42 task outcomes flipped (nearly one-third of the benchmark!). Each solved 36 tasks, with only 15 successes in common. For reference, BF16 with the same Pi harness and thinking mode reached 43.36%. Quantization certainly contributes to trajectory instability: small changes in token probabilities can alter an edit or tool call. That changes what the agent sees next, potentially sending subsequent decisions down a different path. Errors can compound across turns. Separating that from ordinary sampling variability requires repeated experiments. I’d want at least three independent full runs per model with fixed settings before drawing conclusions, and more for a reliable estimate of variability. Too long, too expensive. Don’t trust anyone reporting a quantized model’s accuracy on agentic tasks without also reporting the run-to-run standard deviation.

So many of the activists fighting antisemitism are deliberately pulling the wool over their eyes and choosing not to see what’s right in front of them. That is: Israeli leaders using Jews everywhere to promote Zionist immigration plans and pretending that we stand behind their nationalist plans in the Holy Land. This is not new. Zionists have stirred up, fabricated, and weaponized antisemitism for Jews to “run” to war-ridden Israel and have done everything in their power to silence criticism of the Israeli government by using antisemitism. They are exploiting Jews, enmeshing our identity with a foreign country, and endangering the lives of Jews everywhere. But all the chatter is directed at the people who actually call this out, not at the people who are perpetrating this abominable behavior. Time for a rude awakening here.