Gates 🦞 @sgates

Everyone asking if the new Mac mini can run local models is asking the wrong question. For context, the new Mac mini (M6 or M5 Pro chip) was just announced this morning and is being sold as an AI powerhouse that will cut the amount people are paying @AnthropicAI or @OpenAI every month. To me, the interesting question isn't whether it can run the models. It's whether the math pencils out against a Claude/Codex subscription or an API. Nobody has the new Mac Mini yet, so every speed below is speculation from Apple's spec sheet. Some assumptions: every token a model writes means reading its whole weight file out of memory once, so speed comes down to memory bandwidth. Apple lists: - 153-170GB/s for the M6 - and 307GB/s for the M5 Pro. Nobody has one yet, so I took published runs (see omlx.ai/benchmarks/per…) of these same files on M4 Pro and M5 Max chips and scaled them to the new bandwidth. The Mac build that matters is the $1,299 M6 with 32GB, because a model that doesn't fit in memory gets paged off the SSD at 1-2 tok/s (slower than you type). A 4-bit coding model is 15-20GB before you add the context it's working on, so the $899 16GB device should be ruled out. The M5 Pro with 64GB is $2,700, holds bigger models, and has double the bandwidth. Everything below is a 4-bit build off Hugging Face and fits on the $1,299 M6. In essence, the question isn't what's the best open model. It's what's the best open model that fits in 32GB and still runs at a speed you'd sit through. That rules out the frontier-class ones (DeepSeek V4 Flash, GLM-5.2, Kimi K3 all need 100GB+) and leaves a tier of 27-35B coding models built for exactly this: local, agentic, and quantized to 4-bit by the community within days of release. Here's what I'd pick from, ranked by how well it should run on the cheap box. Candidate Models (Note: Claude Code runs Opus 5 at about 55 tok/s and Sonnet 5 at 70-80): - Qwen3.6-35B-A3B from @Alibaba_Qwen. 20GB at 4-bit, ~35 tok/s. Only 3B of the 35B parameters fire per token, so it's fast on a cheap chip. 73% on SWE-bench Verified. Best coder that also runs well here. - Laguna XS 2.1 from @poolsideai. 19GB at 4-bit, ~35 tok/s; the 3-bit build is 14GB and about 10% faster. Measured at 126 tok/s on an M5 Max, so the mini gets about a quarter of that. Trained inside a coding agent. - Qwen3.8-27B, 11 days old. 17GB at 4-bit, ~10 tok/s on the M6, ~17 on the M5 Pro, measured at 15.5 on an M4 Pro. Dense, so every parameter fires every token. The best local coding agent right now, if you can stand the speed. There's a 2-bit build at 9.8GB that fits the 16GB box; it writes working code with bugs. - Muse Glimmer 30B from @AIatMeta, 2 weeks old. 16GB at 4-bit plus 1.4GB for the vision encoder, ~10 tok/s on the M6, ~18 on the M5 Pro. Also dense, also crawls. Multimodal, made for local agents. - Gemma 4 26B-A4B from @GoogleDeepMind. 14GB at 4-bit, the one here that fits the $1,099 24GB box. ~30 tok/s, and Ollama's draft-model trick nearly doubles that on code. Multimodal, and a weak coder. These are genuinely good models. On agentic coding benchmarks the best of them land about where Claude Sonnet 4.6 was. They are not Opus level though. With the above in mind, here are the takeaways people should be thinking about: 1. Claude Max is $100 or $200 a month and includes Opus 5 and Fable. The $1,299 mini pays for itself in 7 months against the $200 plan, 13 against the $100 plan, and you spend those months on a model two tiers down. The payback is real, it's just for a worse model. 2. A mini running 8 hours a day for 3 years at 35 tok/s puts out about 750 million tokens. That's $1.70 per million tokens of amortized hardware before electricity. Opus 5 charges $25 per million out, so the mini wins by a mile if you'd really burn that many Opus tokens. But @deepseek_ai serves V4 Flash, a stronger model than anything the mini can hold, for $0.66 per million off-peak, and they raised that price last week. Renting the better model is still 2.5x cheaper than owning the worse one. 3. Your code never leaves the room. No rate limits, no five-hour windows, no bill for leaving three agents running overnight, and it still works the day the Claude API goes down. Every one of those is real. None of them shows up as savings. My read: The Mac mini is an excellent privacy box being sold as a cost savings box. If your code can't leave the building, buy the 32GB M6, and run Qwen3.6-35B-A3B. If you want Qwen3.8 or Coder-Next at a usable speed, that's the $2,700 M5 Pro. If you're buying either one to stop paying for Claude, the cheaper move is to keep the $20 Pro plan for the hard problems and point Claude Code at DeepSeek's API for everything else. That's about $40 a month. Model Weights: huggingface.co/Qwen/Qwen3.6-3… huggingface.co/poolside/Lagun… huggingface.co/Qwen/Qwen3.8-2… huggingface.co/meta-models/Mu… huggingface.co/google/gemma-4…

Tim Cook @tim_cook

a day ago

The new Mac mini is here. Small in size. Big on performance. From everyday productivity to all things AI, it can help you do it all.

2K 5K 66K 9.4M 6K

@swill1ams @AnthropicAI @OpenAI Amazing write-up!!!

Introducing Voyage, the first platform to fulfill the promise of AI games. Voyage lets you one-shot an entire world with characters that feel alive, memory that lasts forever, and the structure and challenge needed to make an actually fun game. From the creators of AI Dungeon, the next generation of AI games is here.

@CtrlAltDwayne You can't really put a price tag on owning your own model

@ammaar Thank you so much

@mntruell Please answer as I have 2 subscriptions and not sure what I am doing. x.com/sgates/status/…

Gates 🦞 @sgates

8 hours ago

Day 2. Still hoping @cursor_ai or @xai can clarify. Two-year paying customer here, and I’m getting pretty frustrated that I can’t tell what I purchased or what I’m paying for.

0 0 0 1K 0

@shengzheyao i'm tired of talking to my terminal. I want to talk to my entire operating system.

Is there an agentic app that controls Windows desktop completely by voice? To be able to open programs, search the internet, and do anything just using my voice, not just coding through an app like Codex or Claude.

@adrimvasquezv ahhh Wednesday my new favorite day!.. Love your portal vids!!

@HarperSCarroll This is soooo cool! Very excited for you 🫶

@bot Please answer as I have 2 subscriptions and not sure what I am doing. x.com/sgates/status/…

Gates 🦞 @sgates

8 hours ago

Day 2. Still hoping @cursor_ai or @xai can clarify. Two-year paying customer here, and I’m getting pretty frustrated that I can’t tell what I purchased or what I’m paying for.

0 0 0 1K 0

Your app has AR lenses? Cute. Can you actually interact with them though? 😏👓 OpenLoop can now. AR should be interactable. play.google.com/store/apps/det…

This is amazing!!! I love their product 🫶

superwhisper @superwhisper

7 hours ago

We're making Whisper models free for all users. Starting today, you no longer need a Superwhisper Pro subscription to use them. We’ve also reset Pro usage for everyone. If you’d already reached your limit, you now have 3,000 words to try the latest Pro features. Local,

36 36 588 44K 348

We're making Whisper models free for all users. Starting today, you no longer need a Superwhisper Pro subscription to use them. We’ve also reset Pro usage for everyone. If you’d already reached your limit, you now have 3,000 words to try the latest Pro features. Local, private, and fast voice-to-text. Free for everyone.

@superwhisper

@junwatu

New Mac Studio just dropped. A refurbished 2025 M3 Ultra with 256GB and 16TB is still $15,379. Last year’s machine. Certified used. Still ridiculous. apple.com/shop/product/g…

Day 2. Still hoping @cursor_ai or @xai can clarify. Two-year paying customer here, and I’m getting pretty frustrated that I can’t tell what I purchased or what I’m paying for.

Gates 🦞 @sgates

a day ago

Okay @cursor_ai, genuine question from a 2-year paying customer 👀 I’ve purchased Cursor annually for two consecutive years, with my latest plan costing roughly $200 for the entire year. Today, Ultra is $200 per month. Now that Cursor is part of SpaceX, I’m genuinely curious

keyboard_arrow_left Previous keyboard_arrow_right Next

1 0 0 88 0

@RockstarGames @grok How can I watch or find the video leak!?