cd /news/large-language-models/jamesob-s-guide-to-running-sota-llms… · home topics large-language-models article
[ARTICLE · art-47271] src=github.com ↗ pub= topic=large-language-models verified=true sentiment=· neutral

Jamesob's guide to running SOTA LLMs locally

Jamesob published a guide on building a local system to run state-of-the-art large language models, detailing hardware configurations ranging from $2k to $40k. The setup uses multiple RTX GPUs and PCIe switches for peer-to-peer communication, enabling high-performance inference of models like GLM-5.2-594B and Qwen3.6-27B locally. The guide emphasizes cost-effective VRAM investment over expensive PCIe5/DDR5 hardware.

read9 min views1 publishedJul 3, 2026
Jamesob's guide to running SOTA LLMs locally
Image: source

Note: nothing in this README aside from the tables was written by AI.

Have $2k burning a hole in your pocket and want some local, state-of-the-art machine intelligence?

How about $40k?

If Dario and Altman are giving you heartburn (they should be), read on to figure out how to run this new kind of computing locally.

In this repo you'll find

  • the hardware I use to run SOTA locally,

  • why I bought what and little-known secretsfor configuring it,

  • why I bought what and little-known

  • how I run speech-to-text (STT) locally,

  • ready-to-run configuration for running models I think are good within Docker containers.

Section TL;DR

Base systemGPUsc-payne switch sub-BOMc-payne.comso GPUs talk peer-to-peerGPU mountMaking the switch behaveKernel / GRUB paramsiommu=off

or NCCL hangsACS disableGPU power limitingResultrunners/

GLM-5.2-594B: vLLM docker-compose, DCP4+MTP5, ~80 t/s @ 460k ctxrunners/stt

whisper-large-v3

tools/

: P2P bandwidth/latency benchmarkmeasure-gpu-speed.sh

ResourcesI was lucky/dumb enough to buy 4x RTX Pro 6000s back when they were cheaper. Because RAM is now so expensive, I opted to build a last-gen DDR4 system to host these cards, the parts for which I got off eBay. This allowed me to keep base system cost reasonable while still getting a lot of VRAM.

Another somewhat unusual thing I did was to use PCIe4 switches (from c-payne.com). This allows the GPUs to communicate to one another "directly" at wire speeds during the allreduce step in tensor parallelism, rather than having to send all data through the PCI root complex. The upshot of this is reduced latency between the cards with less of a need for expensive PCIe5 hardware.

Consequently, I'm spending money on VRAM (where it counts) rather than on a PCIe5/DDR5 base system, which is terrifically expensive as of July 2026.

My particular BOM is detailed below.

A great way to go is 2x RTX 3090s for a total of 48GB VRAM total. You can then run Qwen3.6-27B, which is an awesome model.

You can also run SOTA speech-to-text (STT) with whisper-large-v3, which I find very useful. That's the model - you'd then access it with my cross-platform

.

stt

harnessI've found local STT surprisingly useful - and I feel comfortable using it, unlike a hosted equivalent. You can find a ready-to-run config in ./runners/stt that only assumes the presence of ~11GB of VRAM on an Nvidia GPU.

At this price level, you get the next step up in model intelligence. Something pretty close to Claude Opus.

You'd buy 4x RTX 6000 Pros for a total of 384GB of VRAM.

Date Best model My config
2026-07
GLM-5.2-Int8Mix-NVFP4-REAP-594B

Runner configNote: these are my recommendations, but there are other completely valid ways to spend your money. For example, there's probably also some regime where rather than getting 4 rtx6kpros, you allocate most of your money to building out a linked 4x DGX Spark cluster for a total of 512GB VRAM and use that as the slow, big brain to drive Qwen3.7-27b to do the rote tasks quickly.

Here's the hardware I wound up purchasing for the 4x RTX 6000 pro machine.

A modest, last-gen EPYC system purchased in parts almost entirely from eBay.

Component Spec Price
Motherboard ASRock Rack ROMED8-2T (SP3, 7× PCIe 4.0 x16, dual 10GbE) $715
CPU AMD EPYC Milan 7313P (16-core 3.0GHz) $504
RAM 8× 16GB Crucial CT16G4RFD4213 DDR4 ECC RDIMM (128GB total, eBay) $642
CPU Cooler Dynatron T17 SP3 tower, 280W TDP $40
Case AAAWave Sluice V2 open frame $100
PSUs 2× Super Flower 1700W $750
PCIe Switch c-payne Microchip Switchtec PM40100 Gen4 (see sub-BOM below) ~$1,330
Boot NVMe 4TB M.2 $291
Storage NVMe (2x) 8TB M.2 (model weights) $1,200
Fans 3× 120mm PWM $15
Total
$5,587
Component Spec Price
GPUs 4× NVIDIA RTX PRO 6000 Blackwell Workstation (96GB each, 384GB VRAM total)
~$46,000
Part Qty Unit (€) Notes
PCIe gen4 Switch 5× x16 — Microchip Switchtec PM40100 1 1.050 2× SlimSAS 8i upstream, 5× x16 quad-width-spaced downstream, aux x4 SlimSAS, 3× 8-pin EPS power
SlimSAS PCIe gen4 Host Adapter x16 — REDRIVER AIC (DS160PR810) 1 140 Plugs into ROMED8-2T x16 slot, feeds switch upstream
SlimSAS SFF-8654 8i cable — PCIe gen4 2 ~30 Each carries x8; pair = x16 upstream
Total

I had to custom fabricate a wood enclosure for the PCI switch and GPUs, which took about a day.

I found the PCI switch's builtin fan very loud and seemingly useless, so I simply unplugged that from the board.

I save all model weights locally on a ZFS filesystem that's replicated across the two 8TB drives, which is mounted at ~/storage

.

For any model I want to run, I first download the model using

hf download <model-name> --local-dir ~/storage/<model-name>

Once the model weights are cached locally, I have a specific directory for each model that contains a docker-compose.yml

file that cordones off the running of each model to its own Docker container.

You can find these configurations in ./runners/.

Each container mounts in ~/storage/models

in read-only mode to obtain the weights that I've cached locally.

I then use opencode

hosted on a VM on another machine to access the models once they're serving on http://clank.j.co:5000

.

I use a network-internal DNS server to point clank.j.co

to the LLM machine, but you could simply do http://<llm-machine-ip>:5000

too.

I created a VM and clanked up an application that basically just creates a tmux session for each directory within the VM's ~/src

tree, which then runs an opencode

instance that backs up to the inference machine's HTTP API (http://clank.j.co:5000

).

One key to making the opensource models good is tooling them properly; a summary of my skills/

is:

  • camofox, kagi.com API key, and searXNG for web browsing and search,
  • Telegram bot for communication and alerting,
  • a local private Gitea instance for collaborating on source code.

The clanker will either work with me interactively in a session, or can be farmed off to work on Gitea issues and file PRs there.

All this happens in a sandboxed VM where the only communication back to the host system happens via a shared filesystem mount, so the thing can go ham and install whatever it wants.

There was a lot of fiddling with the BIOS in order to make sure the motherboard wasn't downregulating the PCI switch speeds.

Setting Value Why
Chipset Configuration → AMD PCIE Link Width (switch slot)
x16 (was x8/x8)
Bifurcation was splitting the slot; upstream link trained at Gen4 x8. Requires both SlimSAS 8i cables connected (each carries x8).
PCIe Link Speed (switch slot) Gen4 (not Auto)
Blackwell Gen5 devices auto-negotiating down through the Gen4 switch could fail training and fall to Gen1. Forcing Gen4 stabilizes it.
ASPM Disabled
ASPM L1 drops idle links to 2.5GT/s. This turned out to be the explanation for the "Gen1 downgraded" lspci readings — links were actually running Gen4 under load (verified via p2pBandwidthLatencyTest), but disabling ASPM removes the cosmetic scare and any re-train latency.
Re-Size BAR Enabled
Required for full 96GB VRAM BAR exposure and GPU P2P.
SR-IOV Disabled
Bare-metal inference; avoids IOMMU overhead and P2P interference.
Preferred IO Auto
Optionally set Manual → bus 81 (the c-payne switch) for marginal latency gains, but left at Auto — it's a squeeze-more optimization, not a fix, and bus numbers shift after BIOS changes.

Per c-payne's advice, I did reduce the gain to "lvl 3" using his tool, which was probably the most finicky part of the process.

The gain level is going to be a function of how long your SAS connector cables are.

I screwed up and ordered too few of the cables from c-payne directly, so I bought what I thought was the same SAS cable off of Amazon. There was actually a slight difference that was causing issues, and I had to reorder cables - so double-check that you're getting the right stuff!

GRUB_CMDLINE_LINUX="iommu=off amd_iommu=off nomodeset"
sudo update-grub

echo 'options nvidia_uvm uvm_disable_hmm=1' | sudo tee /etc/modprobe.d/uvm.conf
sudo update-initramfs -u

Without iommu=off

, NCCL hangs on multi-GPU P2P.

With ACS enabled (default), P2P traffic gets bounced through the CPU root port instead of staying inside the switch fabric, negating the switch entirely. pcie_acs_override

requires a patched kernel, so we disable via setpci at runtime.

#!/bin/bash
if [ "$EUID" -ne 0 ]; then
  echo "ERROR: must be run as root"
  exit 1
fi

for BDF in $(lspci -d "*:*:*" | awk '{print $1}'); do
  setpci -v -s ${BDF} ECAP_ACS+0x6.w > /dev/null 2>&1
  if [ $? -ne 0 ]; then
    continue
  fi
  echo "Disabling ACS on $(lspci -s ${BDF})"
  setpci -v -s ${BDF} ECAP_ACS+0x6.w=0000
done

Run on every boot via systemd oneshot:

[Unit]
Description=Disable PCIe ACS for GPU P2P
After=multi-user.target

[Service]
Type=oneshot
ExecStart=/usr/local/bin/disable-acs.sh

[Install]
WantedBy=multi-user.target

Verify: lspci -vvv | grep ACSCtl

should show all minus signs, and nvidia-smi topo -m

should show PIX between all four GPUs (not PHB/NODE).

Use ./tools/measure-gpu-speed.sh to measure this easily.

In order to avoid installing a 220V circuit, I (probably unwisely) run this rig on a single 110V circuit, but I power regulate the cards.

Persistence mode + power cap applied at boot via systemd (install-gpu-power-limit.sh):

sudo nvidia-smi -pm 1
sudo nvidia-smi -pl 350    # 350W per GPU (default 600W)

350W/GPU = 1,400W GPU load, sized for the PSU budget. During the interim single-1700W-PSU phase (before the 240V circuit), cards ran at ~260W (4×260 = 1,040W GPUs + ~280W system ≈ 1,320W total).

Verify: nvidia-smi --query-gpu=index,power.limit,power.draw --format=csv

Upstream: Gen4 x16 (~30 GB/s to CPU). P2P through switch: 27.5 GB/s unidirectional / 50.4 GB/s bidirectional, 0.37–0.45 µs latency, i.e. Gen4 line rate. Note: lspci may still show downstream GPU links as "2.5GT/s (downgraded)" at idle if ASPM is active anywhere; this is cosmetic. Links retrain to Gen4 under load.

── more in #large-language-models 4 stories · sorted by recency
── more on @jamesob 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/jamesob-s-guide-to-r…] indexed:0 read:9min 2026-07-03 ·