MiniPC.guru
MINISFORUM MS-S1 MAX

MINISFORUMMini PC reference

MINISFORUM MS-S1 MAX

MINISFORUM MS-S1 MAX uses AMD Ryzen AI Max+ 395 with Radeon 8060S. A compact workstation for large-memory local-AI software or creative workloads with fast wired networking. RAM capacity must be selected at purchase, and the secondary SSD and expansion-card slot have fewer lanes than their physical size may suggest.

Processor
AMD Ryzen AI Max+ 395
Graphics
Radeon 8060S
Memory
128 GB
Storage
2 TB
Configuration being compared128 GB RAM · 2 TB SSDNew
US stores · USD

Prices across stores

MiniPC.guru compares retailer listings. Purchases are completed on the retailer’s website.

View price history

We may earn a commission if you buy through affiliate links.

RetailerAvailabilityListed priceLast checkedRetailer website
Affiliate link
Not confirmedSee at storeCheck expiredView at MINISFORUM
Not confirmedSee at storeCheck expiredView at Amazon

Availability reflects the latest store check. Shipping, tax, coupons and checkout totals are not included; confirm them at the store.

Specifications

Processor, graphics and installed memory/storage describe the selected configuration. Expansion limits and connectivity describe the model; optional hardware is identified separately. References are collected in About this data below.

Processor
AMD Ryzen AI Max+ 395
Graphics
Radeon 8060S
Memory
128 GB
Storage
2 TB
Ports & displays
Two front USB4 40Gbps and two rear USB4 v2 80Gbps Type-C; three USB-A 3.2 Gen 2, two USB-A 2.0, HDMI 2.1 FRL and 3.5 mm combo audio.
Ethernet, Wi-Fi & Bluetooth
Two 10GbE RJ45 ports, Wi-Fi 7 and Bluetooth 5.4.
RAM & storage upgrades
Onboard LPDDR5x-8000, up to 128 GB. Two M.2 2280 SSD slots (up to 8 TB each): one PCIe 4.0 x4 and one PCIe 4.0 x1. Full-length PCIe x16 physical slot runs at PCIe 4.0 x4.
Power supply & power limits
Internal PSU rated up to 320 W. Manufacturer performance/balanced/quiet CPU modes are 130/95/60 W; these settings are not measured wall consumption.
Dimensions & weight
222.1 × 206.3 × 77.1 mm; approximately 2.8 kg.

Video

Plays on YouTube. Nothing is loaded from YouTube until you press play. YouTube

What owners say

Three labs measured this machine at rest and none of them agree. One recorded 46.3 dB(A) from 15 cm and marked its review down for it; another 39 to 41; a third only said it never drops to the room floor. The temperatures diverge the same way, 27 °C on the case at idle in one test against 55 to 60 °C on the vents in another. An owner's fan-control project explains the split: the factory curve reaches only about 40 per cent duty at 93 °C, so the machine runs hot and quiet or cool and loud depending on what the firmware decides, and it restores that curve on every power cycle. A firmware update fixed the idle noise and left the loaded curve alone. Pin the fan at half speed in the firmware and the noise sits flat at 38.5 dB(A) while the chip climbs from 70 °C to nearly 98. The rest is unusually well built for the category: an internal 320 W supply, two 10-gigabit ports, and three M.2 slots you reach by sliding the case off. Two warnings. The second M.2 slot runs at one PCIe lane, not four, and the reviewer who first reported four corrected it. And MINISFORUM's own support says the expansion slot supports neither a discrete nor an external graphics card.

For whom
Large local models that need memory capacity rather than raw speed, in a machine that will also serve a 2U rack over two 10-gigabit ports.
Not for
A graphics card of any kind, and anyone who needs it quiet at idle without going into the firmware first.

Our own summary of 6 published sources, reviewed Sep 17, 2026. We have not tested this Mini PC ourselves; the sources are listed under About this data.

Performance

3 sourced resultstok/s

Same model, quantization, context length, prompt, output length, runtime version and batch size. Record time to first token separately.

Externally reported measurements. MiniPC.guru has not independently reproduced these tests.

16.1 tok/sCommunity benchmark

speed-bench: Strix Halo ROCm results on a Minisforum MS-S1 Max

By alexdelprete · GitHub

Source published
Aug 25, 2026
Test performed
Aug 25, 2026
Added to MiniPC.guru
Sep 17, 2026
Source reviewed
Sep 17, 2026
Tested hardware
AMD Ryzen AI Max+ 395, reported by the contributor as a Minisforum MS-S1 Max · Radeon 8060S, gfx1151, with 124 GiB of GTT exposed through amd_iommu=off, amdgpu.gttsize=126976, ttm.pages_limit=32505856 and ttm.page_pool_size=32505856 · 128 GB LPDDR5X-8000, quad channel · not applicable: the model is fully resident in memory, with no SSD streaming, so token generation does not read from disk. The source does not state the SSD capacity
Software / version / OS
ds4 speed-bench, ROCm backend · ds4 built from main around commit 44e47de with make strix-halo; Ubuntu ROCm 7.1 packages, hipcc 7.1.1, plus the rocWMMA headers named in STRIXHALO.md · Ubuntu 26.04 LTS, kernel 7.0.0-30-generic
Power profile
firmware power mode Performance, about 130 W sustained package limit and a steady 131 W package draw during the run
Test conditions
DeepSeek-V4-Flash IQ2_XXS with w2Q2K, AProj Q8, SExp Q8 and Out Q8, chat v2 imatrix build of 31 July. Command exactly as published in speed-bench/README.md, reading promessi_sposi.txt, context swept from 2048 to 65536 in steps of 2048, 128 tokens generated, machine otherwise idle. This row is context 2048: prefill 153.17 tok/s, generation 16.10 tok/s

Editorial review checks the source and documentation; it does not certify the measurement or mean we tested the device ourselves.

13.3 tok/sCommunity benchmark

speed-bench: Strix Halo ROCm results on a Minisforum MS-S1 Max

By alexdelprete · GitHub

Source published
Aug 25, 2026
Test performed
Aug 25, 2026
Added to MiniPC.guru
Sep 17, 2026
Source reviewed
Sep 17, 2026
Tested hardware
AMD Ryzen AI Max+ 395, reported by the contributor as a Minisforum MS-S1 Max · Radeon 8060S, gfx1151, with 124 GiB of GTT exposed through amd_iommu=off, amdgpu.gttsize=126976, ttm.pages_limit=32505856 and ttm.page_pool_size=32505856 · 128 GB LPDDR5X-8000, quad channel · not applicable: the model is fully resident in memory, with no SSD streaming, so token generation does not read from disk. The source does not state the SSD capacity
Software / version / OS
ds4 speed-bench, ROCm backend · ds4 built from main around commit 44e47de with make strix-halo; Ubuntu ROCm 7.1 packages, hipcc 7.1.1, plus the rocWMMA headers named in STRIXHALO.md · Ubuntu 26.04 LTS, kernel 7.0.0-30-generic
Power profile
firmware power mode Performance, about 130 W sustained package limit and a steady 131 W package draw during the run
Test conditions
DeepSeek-V4-Flash IQ2_XXS with w2Q2K, AProj Q8, SExp Q8 and Out Q8, chat v2 imatrix build of 31 July. Same command and sweep as the context 2048 row. This row is context 65536: prefill 158.32 tok/s, generation 13.30 tok/s, a fall of 17.4 per cent against context 2048

Editorial review checks the source and documentation; it does not certify the measurement or mean we tested the device ourselves.

89.49 tok/sCommunity benchmark

MS-S1 Max on Windows with LM Studio: community report

By bennos1911 · GitHub

Source published
Jun 2, 2026
Test performed
Jun 2, 2026
Added to MiniPC.guru
Sep 17, 2026
Source reviewed
Sep 17, 2026
Tested hardware
AMD Ryzen AI Max+ 395 with Radeon 8060S, 3.00 GHz · AMD Radeon 8060S with 96 GB assigned, set from AMD Adrenalin as a custom profile giving the graphics side its 96 GB maximum and the rest to system memory · 128 GB, of which 31.6 GB remain usable by the operating system under that split · not reported by source: the report does not identify the installed SSD. Token generation does not read from disk once the model is resident
Software / version / OS
LM Studio with the Vulkan llama.cpp runtime · LM Studio 0.4.15 build 2; Vulkan llama.cpp runtime v2.18.0; chipset 8.05.04.516; Adrenalin 26.5.2 · Windows 11 Pro 25H2, build 26200.8457
Power profile
not reported by source: the report does not state the firmware power mode
Test conditions
Qwen3.6-35B-A3B Q4_K_M with the BF16 multimodal projector; n_parallel 4, n_ctx 262144, unified KV cache. Script lm_studio_bench_amd.py with 15 runs per pass and 512 maximum tokens, 45 runs in total. The published value is the mean throughput of 89.49 tok/s; the median of the same runs is 78.53 tok/s and the 512-token prompt rows land around 69 to 70, so the distribution is skewed and the median is the more conservative reading

Editorial review checks the source and documentation; it does not certify the measurement or mean we tested the device ourselves.

Price history

MINISFORUM MS-S1 MAX · 128 GB RAM · 2 TB SSD

Loading checks…

New condition · US stores · USD · Selected configuration only. Prices exclude shipping and tax. Each price stays on the chart until that listing's next check, for at most 36 hours, the same window the store prices use. Out-of-stock, failed and missing checks leave gaps.

About this data11 references

Specs and any benchmark results come from the sources below. We haven't tested this Mini PC ourselves.

Sources

  1. MINISFORUM MS-S1 MAX — manufacturer specifications ↗
    • Manufacturer / specification source
    • Ports & displays · Reviewed Sep 15, 2026
    • Ethernet, Wi-Fi & Bluetooth · Reviewed Sep 15, 2026
    • RAM & storage upgrades · Reviewed Sep 15, 2026
    • Power supply & power limits · Reviewed Sep 15, 2026
    • Dimensions & weight · Reviewed Sep 15, 2026
  2. Selected configuration ↗
    • Selected configuration
  3. MINISFORUM — MS-S1 MAX soldered LPDDR5x explanation ↗
    • RAM & storage upgrades · Reviewed Sep 15, 2026
  4. Minisforum MS-S1 Max im Test: AMD Strix Halo power, 128 GB RAM and Radeon 8060S ↗
    • External benchmark · Reviewed Sep 17, 2026
    • Owner review source · Reviewed Sep 17, 2026
  5. speed-bench: Strix Halo ROCm results on a Minisforum MS-S1 Max ↗
    • External benchmark · Reviewed Sep 17, 2026
  6. MS-S1 Max on Windows with LM Studio: community report ↗
    • External benchmark · Reviewed Sep 17, 2026
  7. MINISFORUM MS-S1 MAX review: inside the chassis ↗
    • Owner review source · Reviewed Sep 17, 2026
  8. MINISFORUM MS-S1 MAX review: power consumption, noise and verdict ↗
    • Owner review source · Reviewed Sep 17, 2026
  9. MINISFORUM MS-S1 MAX review ↗
    • Owner review source · Reviewed Sep 17, 2026
  10. ms-s1-max-fans-control: replacing the factory fan curve on Linux ↗
    • Owner review source · Reviewed Sep 17, 2026
  11. MINISFORUM MS-S1 MAX rear USB4 port instability ↗
    • Owner review source · Reviewed Sep 17, 2026

Related models

Models with related processors, graphics or use cases. Check each model’s specifications and configurations before comparing.

Beelink GTR9 ProAMD Ryzen AI Max+ 395GEEKOM A9 Mega (Max+ 395)AMD Ryzen AI Max+ 395GMKtec EVO-X2AMD Ryzen AI Max+ 395GMKtec EVO-X3AMD Ryzen AI Max+ 395