Disclosure: I represent DigitalCloud.pro. This is a newly provisioned Lite VPS from our own Tokyo (TYO2) location. This comment is a benchmark only — no offer or promo.
Can we at least do GB6? or latest GB V7? we are in 2026, I doubt V4 will give any useful info for some the majoraty...... I mean I can run that in a DDR2 platform and get probably 1000+
@Mason said:
Alright, which one of you fools is this?
It ended up being a handful of Linode IPs hitting the site thousands of times each. I modified how stats are being calculated and am now filtering down to just wget/curl requests (like I should have from the start) so usage stats going forward should be more accurate, but lower than they've been previously since its not including the browser-based requests.
OVH Helsinki — EU-NORTH-LZ-HEL-A, b3-8 (2 vCPU / 8 GB / 50 GB NVMe). This is one of OVH's "Local Zone" regions rather than a normal one, and it doesn't behave like the five I benched a few days ago.
Two things worth flagging before the dump:
It's a different platform underneath. EPYC-Genoa and VM Type: MICROSOFT, where the five normal regions were all EPYC-Milan on KVM.
It breaks a conclusion I posted. In my five-region thread I said the identical 40.0k IOPS at 4k across every region looked like a platform-wide volume quota. Helsinki reads 56.9k. So there is a cap, but it's per-region, not global — correcting that here because the earlier post is where people will find it. Worth noting the large-block numbers went the other way: 2.13 GB/s at 1m is below Frankfurt (5.04) and Mumbai (6.83), so it's a high-IOPS/low-throughput profile rather than a straight upgrade.
Network is genuinely quick to Europe — 7 Gbit/s to Amsterdam against the 489 Mbit/s ceiling every normal region hit — but NYC came back at 142/224 Mbit, which is worse than the shaped boxes were. Port speed and transit quality are clearly two different questions on this one.
GB6 didn't print the scores into the terminal output for some reason; the result link has them: 1764 single / 3144 multi.
Comments
NoAckHosting dedi "2026 Special 1" 800 sek/m
1220 v5 / 16G DDR4 / 250Gb HDD / 100TB@1Gbps
LeaseWeb SIN - base plan (VPS 1)
~S$9.50(?)/mth
Needed something high RAM for once, fits the bill.
GreenCloudVPS 9999 plan in Chicago, migrated to VirtFusion.
Link to previous benchmark in SolusVM.
We accept Karma donations for the last flan. 🍮 affbrr
Ascienth Alpha KVM Standard 3, where RAM is larger than disk due to upgrading.
We accept Karma donations for the last flan. 🍮 affbrr
Lifetime VPS from @crunchbits
Best and friendliest hosts Host-C , Hostbrr aff.
@crunchbits 4c6c100gb with 1TB block storage in the new Synteq's Cloud Platform.
Block Storage fio:
many reguardings!!
very cool!
YABS from the QuickPacket dedicated server from:
and reposted here: https://lowendspirit.com/discussion/comment/239583/#Comment_239583
Notes:
Not sure why disk isn't displaying, probably zfs/raid 1, but it is :
Also, I just haven't configured IPv6 yet, I was assigned a /64.
Network DL is blazing fast:
Pulsed Media: The Eternal Väinämöinen Sale: Seedboxes and storage boxes from €1.99/mo, priced for life. (ref/aff link)
Disclosure: I represent DigitalCloud.pro. This is a newly provisioned Lite VPS from our own Tokyo (TYO2) location. This comment is a benchmark only — no offer or promo.
Configuration: 1 vCPU / 2 GB RAM / 25 GB NVMe / 200 Mbit / Debian 12
Ran YABS immediately after provisioning, with no tuning.
Highlights:
Public Geekbench result: https://browser.geekbench.com/v4/cpu/19845830
Full YABS output:
Guys, just wondering,
I get that you do all this tests and post them, yet, if there is no agreement on GB version, it is pretty misleading
I was like WTF?
Then I saw it was GB V4,
Can we at least do GB6? or latest GB V7? we are in 2026, I doubt V4 will give any useful info for some the majoraty...... I mean I can run that in a DDR2 platform and get probably 1000+
Just my 2c on this.
Host-C | Storage by Design | AS211462
“If it can’t guarantee behavior under load, it doesn’t belong in production.”
KS-3, got 2x 4tb instead of 2x 2tb
https://hammervm.com YABS:
Plastic Limited Stock
1 Cores [E5-2640v4]
2 GB DDR4 ECC RAM
20 GB HDD
100 Mbit/s (Fair Usage)
£2.99/month VPS
`# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
Yet-Another-Bench-Script
v2026-07-24
https://github.com/masonr/yet-another-bench-script
## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ##
Fri Aug 28 21:15:27 UTC 2026
Basic System Information:
Uptime : 0 days, 0 hours, 0 minutes
Processor : Intel(R) Xeon(R) CPU E5-2640 v4 @ 2.40GHz
CPU cores : 1 @ 2397.222 MHz
AES-NI : ✔ Enabled
VM-x/AMD-V : ✔ Enabled
RAM : 1.9 GiB
Swap : 0.0 KiB
Disk : 19.6 GiB
Distro : Debian GNU/Linux 12 (bookworm)
Kernel : 6.1.0-48-cloud-amd64
VM Type : KVM
IPv4/IPv6 : ✔ Online / ✔ Online
IPv6 Network Information:
ISP : Privex Inc.
ASN : AS210083 Privex Inc.
Host : Privex Sweden
Location : Stockholm, Stockholm County (AB)
Country : Sweden
fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/sda1):
iperf3 Network Speed Tests (IPv4):
iperf3 Network Speed Tests (IPv6):
Geekbench 6 Benchmark Test:
Test | Value
|
Single Core |
Multi Core |
Full Test | https://browser.geekbench.com/v6/cpu/19081204
YABS completed in 19 min 56 sec
`
link please
Looks like the 18 Eur/yr plan is OOS. There's still a couple of 9 Eur/yr
https://loclix.io/store/let-deals
Best and friendliest hosts Host-C , Hostbrr aff.
Alright, which one of you fools is this?
yabs daddy
@AlteredParadox for sure
onidel - 🇦🇺🇳🇱🇸🇬🇻🇳🇺🇸 - High Availability - Hourly Billing - API support - BGP/BYOIP - SEV-SNP
It ended up being a handful of Linode IPs hitting the site thousands of times each. I modified how stats are being calculated and am now filtering down to just wget/curl requests (like I should have from the start) so usage stats going forward should be more accurate, but lower than they've been previously since its not including the browser-based requests.
yabs daddy
OVH Helsinki — EU-NORTH-LZ-HEL-A, b3-8 (2 vCPU / 8 GB / 50 GB NVMe). This is one of OVH's "Local Zone" regions rather than a normal one, and it doesn't behave like the five I benched a few days ago.
Two things worth flagging before the dump:
It's a different platform underneath. EPYC-Genoa and VM Type: MICROSOFT, where the five normal regions were all EPYC-Milan on KVM.
It breaks a conclusion I posted. In my five-region thread I said the identical 40.0k IOPS at 4k across every region looked like a platform-wide volume quota. Helsinki reads 56.9k. So there is a cap, but it's per-region, not global — correcting that here because the earlier post is where people will find it. Worth noting the large-block numbers went the other way: 2.13 GB/s at 1m is below Frankfurt (5.04) and Mumbai (6.83), so it's a high-IOPS/low-throughput profile rather than a straight upgrade.
Network is genuinely quick to Europe — 7 Gbit/s to Amsterdam against the 489 Mbit/s ceiling every normal region hit — but NYC came back at 142/224 Mbit, which is worse than the shaped boxes were. Port speed and transit quality are clearly two different questions on this one.
GB6 didn't print the scores into the terminal output for some reason; the result link has them: 1764 single / 3144 multi.
## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ##
Yet-Another-Bench-Script
v2026-07-24
https://github.com/masonr/yet-another-bench-script
## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ##
Sun Sep 13 19:14:52 UTC 2026
Basic System Information:
Uptime : 0 days, 0 hours, 7 minutes
Processor : AMD EPYC-Genoa Processor
CPU cores : 2 @ 3095.334 MHz
AES-NI : ✔ Enabled
VM-x/AMD-V : ✔ Enabled
RAM : 7.8 GiB
Swap : 0.0 KiB
Disk : 48.4 GiB
Distro : Ubuntu 24.04.4 LTS
Kernel : 6.8.0-106-generic
VM Type : MICROSOFT
IPv4/IPv6 : ✔ Online / ✔ Online
IPv6 Network Information:
ISP : OVH SAS
ASN : AS16276 OVH SAS
Host : OVH HOSTING OY
Location : Helsinki, Uusimaa (18)
Country : Finland
fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/sda1):
iperf3 Network Speed Tests (IPv4):
iperf3 Network Speed Tests (IPv6):
Geekbench 6 Benchmark Test:
Test | Value
|
Single Core |
Multi Core |
Full Test | https://browser.geekbench.com/v6/cpu/19166833
YABS completed in 13 min 10 sec
Usual caveats: one instance, one run, no long-term observation, and a Local Zone says nothing about OVH's normal regions.
Raw output for the earlier five regions (Frankfurt / London / Milan / Warsaw / Mumbai, six runs) is in the same gist: https://gist.github.com/vpnai2002/7f63df4a81da5a2c6a67dafbd35b028a