They say things like "no matter what it takes, no matter how many of us die, we must be free again, this time we will win against the terrorist regime" (paraphrased).

-

_(Sorry for the joke. I know nothing about you, it was just a cheap one-liner I felt can be shared with this disclaimer. Much love.)_

You said: "No it won't. It really, really wont."

With the explosive development of LLMs and their abilities, it seems your point of view is probably the hopeful one while the other poster has the realistic one.

It seems that you simply can't say anything about what LLMs will not be able to do. Especially when you try to use current "AI slop" as your main reason, which is being more and more eradicated.

I have a MacBook Pro M1 Max w/64 GB RAM, and a Mac Studio M3 Ultra w/96 GB RAM. What do you think is possible to run on these? Just curious before I really try it out.

Can you share more info about quants or whatever is relevant? That's super interesting, since it's such a capable model.

I'm using LM Studio now for ease of use and simple logging/viewing of previous conversations. Later I'm gonna use my own custom local LLM system on the Mac Studio, probably orchestrated by LangChain and running models with llama.cpp.

My goal has all the time been to use them in ensembles in order to reduce model biases. The same principle has just now been introduced as a feature called "model council" in Perplexity Max: https://www.perplexity.ai/hub/blog/introducing-model-council

Chats will be stored in and recalled from a PostgreSQL database with extensions for vectors (pgvector) and graph (Apache AGE).

For both sets of tests below, MLX was used when available, but ultimately ran at almost the same speed as GGUF.

I hope this information helps someone!

/////////

Mac Studio M3 Ultra (default w/96 GB RAM, 1 TB SSD, 28C CPU, 60C GPU):

• Gemma 3 27B (Q4_K_M): ~30 tok/s, TTFT ~0.52 s

• GPT-OSS 20B: ~150 tok/s

• GPT-OSS 120B: ~23 tok/s, TTFT ~2.3 s

• Qwen3 14B (Q6_K): ~47 tok/s, TTFT ~0.35 s

(GPT-OSS quants and 20B TTFT info not available anymore)

//////////

MacBook Pro M1 Max 16.2" (64 GB RAM, 2 TB SSD, 10C CPU, 32C GPU):

• Gemma 3 1B (Q4_K): ~85.7 tok/s, TTFT ~0.39 s

• Gemma 3 27B (Q8_0): ~7.5 tok/s, TTFT ~3.11 s

• GPT-OSS 20B (8bit): ~38.4 tok/s, TTFT ~21.15 s

• LFM2 1.2B: ~119.9 tok/s, TTFT ~0.57 s

• LFM2 2.6B (Q6_K): ~69.3 tok/s, TTFT ~0.14 s

• Olmo 3 32B Think: ~11.0 tok/s, TTFT ~22.12 s

1. The “Double” Military-Industrial Complex (with numbers) The article’s “economy of occupation” frame is incomplete: Gaza is a proxy-war zone where both blocs run industrial supply chains.

Western/Israeli MIC: Quincy Institute documents “at least $21.7 billion” in US military aid since Oct 7, 2023, funding Iron Beam lasers, JDAM kits, and munitions replenishment.[2] This is state-scale industrial output, not incidental corporate profiteering.

Iran-linked proxy MIC: Iran provides Hamas $350 million annually (2023 Israeli security source) and Hezbollah $700+ million/year, but has shifted from direct shipments to “broker of military-industrial knowledge,” transferring production blueprints for indigenous missile/UAV factories.[3][4] Alma Research notes this “hybrid doctrine” lets proxies manufacture locally, reducing interdiction risk.[4] Ignoring this material capacity misrepresents the war as asymmetric in only one direction.

2. “Genocide” is used by major bodies but remains legally indeterminate The article treats the label as settled. Empirically, it is not.

Who uses it: Amnesty International (Dec 2024) concluded there is “sufficient basis” to say Israel is committing genocide.[5] UN special rapporteurs have adopted the term.

Why it’s contested: The 1948 Convention requires “intent to destroy, in whole or in part, a national, ethnical, racial or religious group.”[6] The core dispute is inferring intent from conduct. NPR summarises: “it’s not always clear if they mean Hamas or Gazans.”[7] The ICJ’s final judgment on South Africa v. Israel is expected late 2027 or early 2028.[8] Until then, presenting the charge as fact rather than a plausible but unproven legal claim is premature.

3. Albanese’s criticism is methodological, not personal UN Watch’s legal analysis notes her June 2025 report uses “genocide” 57 times while “Hamas” and “terrorism” appear zero times (excluding footnotes).[1] Four governments (US, France, Germany, Canada) have condemned her approach.[9] The Special Rapporteur mandate itself is anomalous: it is the only HRC mandate that is indefinite (“until the end of the Israeli occupation”) and examines only Israeli violations, systematically excluding Palestinian armed groups.[10] This isn’t about “standing with the oppressed”; it’s about whether a mandate designed for activism can produce impartial analysis.

Bottom line: HN should discuss the political economy of proxy wars and the failure of international law to handle non-state industrialised conflict, not personality-driven morality tales.

SOURCES: [1] Georgetown University drops UN's Albanese due to US sanctions https://www.timesofisrael.com/georgetown-university-drops-un... [2] U.S. Military Aid and Arms Transfers to Israel, October 2023 https://quincyinst.org/research/u-s-military-aid-and-arms-tr... [3] Iranian support for Hamas - Wikipedia https://en.wikipedia.org/wiki/Iranian_support_for_Hamas [4] Hezbollah – Independent Weapons Production, a Hybrid Doctrine in ... https://israel-alma.org/hezbollah-independent-weapons-produc... [5] Amnesty concludes Israel is committing genocide in Gaza https://www.amnesty.org/en/latest/news/2024/12/amnesty-inter... [6] [PDF] Convention on the Prevention and Punishment of the Crime of ... https://www.un.org/en/genocideprevention/documents/atrocity-... [7] A question of intent: Is what's happening in Gaza genocide? - WGBH https://www.wgbh.org/news/2025-09-25/a-question-of-intent-is... [8] Whatever happened to South Africa's case at the ICJ? https://www.middleeasteye.net/explainers/israels-genocide-ga... [9] UN Watch Refutes Biased New Report by Francesca ... https://unwatch.org/un-watch-refutes-biased-new-report-by-fr... [10] UN Must Intervene on Flawed Special Procedure Mandate https://ngo-monitor.org/submissions/submission-to-unhrc-57th...

The beauty with this availability is that any model you switch to can read the whole thread, so it's able to critique and augment the answers from other models before it. I've done this for ages with the various OpenAI models inside ChatGPT, and now I can do the same with all these SOTA thinking models.

To my surprise Kimi K2 is quite sharp, and often finds errors or omissions in the thinking and analyses of its colleagues. Now I always include it in these ensembles, usually at the end to judge the preceding models and add its own "The Tenth Man" angle.

That and being able to mix my own uploaded tracks with online music releases into a curated collection almost made it a viable contender to my local iTunes collection.

And then... they just removed it forever. Bastards.

I might be wrong, though. But this was the initial conclusion I arrived at when I was researching whether to buy an iPhone 17, iPhone 15 Pro (used) or Android phone. Only the last option would probably hurt Apple directly. And only a liiiiiittle.