All Activity

I will Upload that 8B RX 5600xT bench result in a few hours. Modell 4B was 52.1. / 100.

Update, with the suggested model, qwen 3.6 35b a3b Window server crashed and I had to reboot my machine. If I try to do a benchmark with it, the system starts hanging for a few seconds, starts responding and hangs again, I can see the gpu being spiked but seems like it's not able to complete it and I have to stop it. Regarding feedback, the webpage console is in Spanish even selecting English as main language and my 6900xt is recognized as a 6700. If you need help translating the application to Italian, I can help.

A new video presentation of the Tahoe Drive Icons Pack is now available! Many thanks to @kaoskinkae for taking the time to create and publish this video. The video, especially for our Spanish-speaking friends, provides a nice overview of the project, the different icon collections included in the pack and demonstrates how the icons can be applied to drive volumes. Please note: Contrary to the video, ICNS files are recommended for macOS. Icons can be applied directly via Finder using drag & drop: Right click the volume → Get Info → drag the desired icon onto the small volume icon in the upper left corner of the Info window. For Windows and Linux volumes, write access to the respective filesystem is required (for example via Paragon NTFS or Paragon extFS). Desktop icon size can be adjusted via: Right click Desktop → Show View Options → Icon Size. Video: Tahoe Drive Icons Pack: https://github.com/kgp-macPro/Tahoe-Drive-Icons-Pack Many thanks again to @kaoskinkae for supporting the project! 😊🍏

Here my bench P.s. It's a real MacPro not an Hackintosh

Nice, thanks for testing! That RX 5600 XT result is genuinely great. Just a heads-up for the comparison: you ran Qwen3-4B, while my 101/57 numbers are for the bigger Qwen3-8B — so not quite the same test. Your prompt speed basically matches my 6700 XT; generation is a bit lower because it's bandwidth-bound and the 5600 XT has less memory bandwidth (no Infinity Cache). It'll fit the 8B fine too (4.7 GB) if you want a direct apples-to-apples run — and your DDR5 + 12400F will really shine on the bigger MoE models. The MacBook's RX 560X being slow is expected — that's an old Polaris chip with very little VRAM and bandwidth, so generation falls off a cliff (the model can't really stay resident on the GPU). Prompt still looks OK because it's batched, but ~1 t/s gen is just the card showing its age. The 5600 XT is the one to use. For that 4B model i got 68 t/s | 146 t/s on the RX 6700 XT If you grab the 8B and share the numbers, I'd love to add the RX 5600 XT as a tested card. Appreciate the report!

Hi, worked on my Hackintosh with RX 5600 XT - I5 12400F 4,8 GHZ OC DDR5 ( and Macbook Pro RX560x - but very slow) The smallest LLM gave 52.1 / 100.0 in the benchmark . close to your 6700XT? Mobile RX 560X on MacbookPro

@mnfesq, VoodooHDA was working with Nvidia HDMI. New changes are intended for AMD HDMI if not break old functionality. But first you you have to do nvidia framebuffer working.

Hi all, im sharing a project that might be useful to anyone running an AMD GPU on a Hackintosh. The problem: local‑LLM tooling on macOS targets Apple Silicon. On Intel Macs with discrete AMD GPUs, stock llama.cpp under Metal produces corrupted output and is painfully slow over PCIe. ToshLLM is a native SwiftUI app (pure Swift Package Manager, no external deps) that bundles llama.cpp built with AMD‑specific patches and wraps it in a real GUI: Correct Metal output on AMD dGPUs at full speed Qwen3‑8B Q4: ~101 t/s prompt / ~57 t/s generation Qwen3.6‑35B‑A3B (MoE, hybrid offload): ~123 t/s / ~18.6 t/s, up to ~25.7 t/s with MTP Native chat (Markdown, code copy, file attachments), model manager with per‑model VRAM/RAM estimates, automatic MoE CPU‑offload, MTP speculative decoding, dual engines (official + TurboQuant for 100k+ ctx), built‑in benchmarks, OpenAI‑compatible API, bilingual EN/ES New macOS 26 “Tahoe” Liquid Glass interface (degrades to translucent materials on macOS 14/15) Hardware: developed on RX 6700 XT 12 GB + [NootRX](https://github.com/ChefKissInc/NootRX); runs on any working Metal setup Its beta. DMGs aren’t notarized yet (first launch needs “Open Anyway” or `xattr -dr com.apple.quarantine`). The AMD patches live in the repo (`patches/`), so you can build from source too. License: GPL‑3.0. Repo, source and DMG releases: Link to Github Project Would love testing reports from other macOS-supported AMD cards: RDNA 1 (RX 5500/5600/5700), RDNA 2 (RX 6600/6700/6800/6900), and older Polaris/Vega.

Installed Ok On My Mac mini With 256GB SSD Running 26.5.1, iPad Mini, Not Enough Free Space To Install Itt On My iPhone12 mini Serve Me Right For Getting A 64GB One Attempted To Install To My Main System, 512GB SSD, NO ROOM I Delete Some Stuff And Still Shows The Same Amount Free, WTF Formatted My External And Will Try Again With It

It is safe, but the sequence of actions depends on whether OCLP was used. For sound you need either AppleHDA.kext or VoodooHDA.kext

I've been following this thread and trying some of the fixes offered but without success. Here's what is going on with my new laptop (see specs in my signature): I started out with MaLd0n's EFI from Olarila but sleep didn't work so I extracted my DSDT using Clover and modified all _PRW methods from 6D to 09 as suggested here: https://github.com/grvsh02/A-guide-to-completely-fix-sleep-wake-issues-on-hackintosh-laptops. Now, sleep works great but, when my laptop wakes from sleep I lose audio in my internal speakers but audio still works fine using bluetooth headphones/speaker or the pin plug for headphones. Also, this laptop has IGPU (Intel UHD 630) and DGPU (Nvidia RX3000) and I believe that digital audio/HDMI is linked to the DGPU. Of course, I'm using VoodooHDA or I wouldn't be posting this here. I'm thinking that a modification of my DSDT or an SSDT might make the internal speakers work after waking from sleep. As for HDMI, 2 laptops ago, I had the same or similar problem and was able to fix it by injecting the Nvidia device even though it had no driver and HDMI would then be handled by the IGPU. So far, that hasn't worked but I probably injected the wrong information and am still working on it. But I was intrigued by MaLd0n's injection of ATI DGPU properties into VoodooHDA and wondered whether that could work for Nvidia as well. Any insights would be appreciated.

Well I Downloaded tthe Beta App And Installed It To An External Drive, Ha, Don't BootI I Think It Wants To Be Installed To My 26.51 Internal SSD Wont't Do That, But Will Have A Go Installiing It To My Mac mini

Just A Place Holder Till I Install The Beta

Hello volks I am on 26.0 i plained to update to 26.5.1 and i have no sound on board Which kext do i need ? its safe to update ?

For all Intel Wi-Fi users following the ongoing Tahoe testing: I have opened AirportItlwm issue #1062 to investigate the current AWDL situation under macOS Tahoe. Wi-Fi, Bluetooth, AirPlay, Screen Mirroring and Handoff appear to be working with the current AirportItlwm/OCLP approach, but AirDrop, Personal Hotspot and Continuity Camera remain unavailable due to the apparent absence of awdl0. Additional test results, logs and feedback from AX2xx, BE200 and older Intel chipsets are highly appreciated. GitHub issue: https://github.com/OpenIntelWireless/itlwm/issues/1062

Alright cool, thanks for the info, i gone into installing 26.6 and 26.5.1 both of them broke the sleep wake, i switch back to 26.5 alone and the sleep wake functions again. Seems going above that version, it totally broke the sleep wakes.

From what I can tell, this release isn’t about improvements—it’s about dropping Intel. I installed it on an external NVMe drive connected to my Mac mini M4, and I can really notice a difference in speed. I'll give them that. As for new features, there’s practically nothing worth mentioning. New Siri? Big deal. I haven’t tested everything yet, but so far, nothing really stands out.

Hey guys! Great news if you're a fan of AOE 2 DE, now it's available at Steam for macOS. But there's a catch, only for Mx processors. A nice thing to see, it's that even the A18 Pro (Macbook Neo) can run it 😁 PS: If you already buyed the Windows version, you don't need to buy it again.

Very ugly names. I'l try check it.

Hi guys, I uninstalled Golden Gate on my M4 Pro because a few bugs are interfering with two of my essential professional activities. First, Firefox no longer receives passwords directly; authentication now only works via code and through Safari. Second, VS Code with the Chat for Copilot extension is unable to sign in to GitHub. Additionally, compiling older apps such as Clover, which require deployment targets as far back as macOS 10.6, is no longer possible, as it only accepts macOS 12 as the minimum target. For these reasons, I have already erased Golden Gate and reinstalled Tahoe. Good luck to the new users testing it. In my opinion, however, only a few things have changed, and nothing particularly extraordinary.

Ok now its HackMacintosh

I've tested the latest version of @MaLd0n and it works fine HDMI/DP is not working

The shrink will come with macOS 28, when Rosetta 2 and all remaining x86 code are removed—though probably not back down to 5 GB.

Is it the LayoutId that makes it display as Internal Speakers?

He's using VirtualSMC.kext without boot-arg vsmcgen=1. As far as I know, that won't work. He can add the boot-arg or switch to FakeSMC. Haven't heard from him for a while, so maybe he really is lost.