

🚀 Unlock your device’s hidden GPU power and speed — don’t get left behind!
The Chenyang Oculink SFF-8612 to PCI-E 3.0/4.0 M.2 M-Key adapter enables ultra-fast 32 Gbps data transfer by transmitting PCIe 4x NVMe signals, allowing seamless connection of external GPUs and SSDs to compatible laptops and mini PCs. Designed for high performance and broad OS compatibility, it’s the essential upgrade for professionals and gamers seeking to boost graphics and storage capabilities externally.







| ASIN | B0BZW1G87R |
| Best Sellers Rank | #206 in SATA Cables |
| Brand | chenyang |
| Compatible Devices | Laptop |
| Connector Type | Oculink |
| Customer Reviews | 4.3 4.3 out of 5 stars (162) |
| Finish | Pci-e |
| Item Dimensions | 3.94 x 3.94 x 0.39 inches |
| Item Weight | 0.09 Kilograms |
| Manufacturer | CY |
| Model Number | CY-SF-014+SF-061 |
| Number of Items | 1 |
| Number of Ports | 1 |
| Package Quantity | 1 |
| Power Plug Type | No Plug |
| Unit Count | 1.0 Count |
H**E
Works perfectly
Hooked up a 750w psu and a 750ti to my kids dell optiplex aio 2nd m2 port and it works like a charm. Integrated graphics are a no go when it comes to gaming, even an old 750ti stomps hd 630 Intel Totally worth it, and it was a fun process installing and building an enclosure.
N**.
Works! Keep in mind cable orientation.
It works! The oculink connector to your computer comes out straight up/out from the board, so keep in mind the orientation of your m.2 connector. For a laptop, this means it comes straight out the bottom, not the side or back.
G**.
Works with more than GPD stuff! Full 3.0 4x!
Works with a random mini PC running a 5800H. I added an RX6600 to the 16x slot, plugged in a 500w bronze PSU, got everything all wired up, clicked on the tiny power switch, and then powered up the mini PC. Worked first time! Took a minute or so, the screen blanked out as the drivers where applied to the RX6600 that I plugged in. Ran a super solid 3Dmark timespy score, but in actual games you may see more of a bottleneck with the 3.0 4x slot if its a higher end GPU. Works for my use case, cheap too!
H**R
Works great!
I received this today, briefly tested it and it works great! I briefly tested an old GTX 970 on this pcie x16 board with the included m.2 adapter and a different m.2 adapter, they both worked. Also, tested with this with an m.2 to m.2 extension cable. This did NOT work. I did not check how many pci lanes are actually running, but I will update here once I do. And I just threw this together rather quickly (zip ties and scrap metal bottom). I have ordered an external gpu bracket to hold it all better.
D**J
Quality concerns, defective and missing item
Has Quality issues. Was not boxed properly (loose) and arrived damaged with broken capacitor. Was missing the Occulink cable. Do not recommend!
F**2
Pretty good
Let's drive the very first important point here; a WIN MAX 2 is NOT the only device you are allowed to use for this. I bought this for my GPD WIN MINI, which has an oculink port, and this has worked fine. Personally, for $40 dollars the price is very reasonable, especially considering that oculink cables can be expensive. The only gripe I have with this is the board; there is no supporting bracket or piece of anything underneath the actual board to give it some strength. When I connect my graphics card to it, I feel as though I'm going to snap the board because of how thin it is. Also take into consideration that the board does not have those locking clips to give an audible sound that the gfx is connected, you have to provide a little force to make sure it's connected with this product. A little scary, worried that I am going to break the actual board trying to connect or disconnect, but time will tell. Overall this is a must buy if you have oculink and want to use an EGPU. I pair this with a 6700xt to a MINI and get decent performance. For the most recent game example, Hellblade 2 gets me around 68fps on high, 1080P, 120hz. Compare this to TB4, I get around 54fps overall. Oculink definitely gives a nice little bump .
R**N
Works as expected with auto psu switching
Running pcie4.0 x4 with my 3080ti and it works fine. It has automatic (and manual) switching of the psu which is really good for power saving (auto off in suspend/sleep and when host is off). The pins are quite exposed though so be careful with that.
W**S
Worked with my 1070ti and Aoostar Gem10 mini-PC.
This thing works. I attached it to a Aoostar Gem10 through the Oculink port and powered it with a standard PC power supply. This powered my Nvidia 1070ti with no issues and was able to deliver full video support to my mini-PC. I have not tested the NVMe adapter, but i'm sure that works as expected. I didn't really need it for that use case. I bought this for $50 so I would say it was the best value for the money.
L**N
Bonsoir, Merci beaucoup pour la rapidité de livraison, le produit me paraît conforme, mais comme c'est pour un cadeau, je ne sais pas s'il fonctionne comme prévu.
K**I
intel arc a750にて動作しました。 ミニPCのm.2がpcie3.0だったこともありベンチマークスコアとしては直挿し4.0×16より15%〜20%ほど下落しています。 特に不具合もなく相性問題などeGPUではよく話題にあがるものも発生せず安定しています。 ただ私の構成では電源連動はしなかったです。 m.2スロットには付属のカードではなく、別のoculinkメス端子と止めネジがついたものを使用しケースに固定していました。
H**E
Product works with my gpd win 4 2024, for people struggling to get it to work here's some tips 1. There's a microscopic on and off switch on the board, make sure it's switched on ( i found it after scratching my head for ages ) 2. If you are trying to connect an nvidia card and get code 43 look up nando4 fix You are welcome
O**E
This worked just fine. Please consider if the ATX connector direction is compatible with the case you might use. I had to desolder and re-orient mine, which is perfectly doable.
T**G
Really happy with this purchase, it comes with the oculink cable and wifi slot to oculink connector but I bought another one with extender through the nvme slot. Much better than the noisy overpriced Aoostar. The build is combined with RTX 5070 and Corsair RM750e modular PSU for a clean build. Zero noise build, the fan only goes up if temperature passes 50 degrees Celsius. The GPU is turn on and off according to the laptop power state. Plug in, Power up, Install latest Nvidia Driver, Gigabyte control center, code 43 script. I write low level CUDA for a living, so here’s some extra info: - nvidia-smi shows 2 GPUs (my original laptop GPU rtx 3050Ti and this one) - 2 GPUs can communicate through PCIe, I tested a NCCL code for vector addition on 2 GPUs and it worked. - CUDA Code can compile with sm_120, and run no problem. Pytorch recognizes 2 GPUs and set the 5070 as GPU 0. Nsight compute works with debugging and uses 5070 if the code is compiled with sm_120.
TrustPilot
3 周前
2 周前