Update 10/11: This deal is now expired. HPDEAL25 returns "Promotion Coupon_HPDEAL25_WK49-728589935 is temporarily not available at this time" and no discount applies. For the record, the coupon was verified working on the evening of 10/10, when 25% off applied correctly to all three configs at checkout. HP appears to have pulled the code overnight.
HP's new OmniBook X 14 with the NVIDIA RTX Spark N1X superchip (Grace ARM CPU + Blackwell GPU, up to 128GB unified LPDDR5x memory) is 25% off at HP.com with code HPDEAL25. These are the lowest RTX Spark N1X laptop prices we have seen anywhere.
Deal Instructions:
1. Visit the HP OmniBook X 14-KS000 product page:
https://www.hp.com/us-en/shop/pdp...-cb1l6av-1 and click Customize and Buy
2. Pick your config: 32GB ($2,999.99), 64GB ($3,999.99), or 128GB ($5,109.99), all with 1TB SSD
3. Add to cart and apply promo code HPDEAL25 for 25% off
4. Checkout. Shipping is free. This is a custom build with an estimated ship date of November 6, 2026.
Prices after code HPDEAL25:
- 32GB / 1TB SSD: $2,249.99 (reg. $2,999.99)
- 64GB / 1TB SSD: $2,999.99 (reg. $3,999.99)
- 128GB / 1TB SSD: $3,832.49 (reg. $5,109.99)
About this deal:
- Our research indicates the 32GB config is $1,150.00 less (34% savings) than the next best price we found: $3,399.99 at Micro Center (preorder for October 16, in-store pickup only). The 64GB config is $1,380.00 less (32% savings) than Micro Center's $4,379.99.
- The 128GB config has no Micro Center equivalent. For comparison, NVIDIA's own 128GB DGX Spark desktop now lists at $4,699 and runs Linux only. This is a full Windows laptop with a 14-inch 3K OLED 120Hz touchscreen for less.
- Why 128GB unified memory matters: the entire memory pool is shared between CPU and GPU, so it can hold 70B-class AI models fully in memory for local inference and fine-tuning. Nothing else at this price offers that.
About this product:
- NVIDIA RTX Spark N1X superchip: 20-core Grace ARM CPU with 6144-core Blackwell GPU (32GB config uses the 18-core / 5120-core variant), up to 1 Petaflop FP4 AI performance
- 14-inch 3K OLED 120Hz touchscreen, LPDDR5x unified memory, Windows 11 Home
- Please note: RTX Spark is ARM-based, so this runs Windows 11 on ARM. Check ARM compatibility for your must-have apps before buying. Code HPDEAL25 is limited to 2 discounted items per order.
Join The Conversation
Share your experience with the Slickdeals community
Share information with the community. Please follow our Community Guidelines and be kind!
5 Comments
Sign up for a Slickdeals account to remove this ad.
Our community has rated this post as helpful. If you agree, why not thank
For those looking to use this for AI.
Keep in mind that this has very small memory bandwidth just like the DGX Spark
Memory Bandwidth: Operates on a 256-bit memory interface with LPDDR5x-9400 / effective speeds yielding roughly 273 to 300 GB/s of memory bandwidth. Token generation (inference speed) for local LLMs will be heavily bound by this ~273–300 GB/s bandwidth limit.
a regular 5090 LAPTOP has 900GB/s bandwidth.
M5 Max 40GPUs has 614GB/s.
So if you have the same 128GB bandwidth, your inference speed on this guy is 1/3 of a 5090 (memory constrained) and 1/2 of the M5 Max (since RTX laptop and M5 Max have unified memory)
Gaming Equivalency: Nvidia pitches the 6,144-core GPU variant as delivering gaming performance roughly comparable to a mobile GeForce RTX 5070 (or matching the core count of the desktop RTX 5070). It targets high-refresh-rate 1440p gaming utilizing DLSS 4.5.
Geekbench6: 2,570 points in Geekbench 6 single-core and around 23,126 points in Geekbench 6 multi-core tests. https://www.tomshardwar
it is a pretty low pricer to get up to 128GB RAM though compared to all other devices!
If the machine wasn't so god damn ugly looking I might try one.
but then again I didn't need my 128GB Macbook M5 Max so I don't think I'll use up all of the memory in this RTX ARM laptop either. I wonder if the software suite still sucks ass for Windows ARM. lol. (for me - I'm sure many will say it works just fine but those people so far never used snapdragon + windows arm-chair qb)
I thought someone said Linux may actually get support since Nvidia is behind this chip instead Snapdragon -- which appears to be dying even slower death given RTX Spark cpu is out.
Promotion "Coupon_HPDEAL25_WK49-728589935" is temporarily not available at this time.
maybe it's not activated yet. It is not on the official coupon thread either. I did wonder how the OP get this coupon.
For those looking to use this for AI.
Keep in mind that this has very small memory bandwidth just like the DGX Spark
Memory Bandwidth: Operates on a 256-bit memory interface with LPDDR5x-9400 / effective speeds yielding roughly 273 to 300 GB/s of memory bandwidth. Token generation (inference speed) for local LLMs will be heavily bound by this ~273–300 GB/s bandwidth limit.
a regular 5090 LAPTOP has 900GB/s bandwidth.
M5 Max 40GPUs has 614GB/s.
So if you have the same 128GB bandwidth, your inference speed on this guy is 1/3 of a 5090 (memory constrained) and 1/2 of the M5 Max (since RTX laptop and M5 Max have unified memory)
Gaming Equivalency: Nvidia pitches the 6,144-core GPU variant as delivering gaming performance roughly comparable to a mobile GeForce RTX 5070 (or matching the core count of the desktop RTX 5070). It targets high-refresh-rate 1440p gaming utilizing DLSS 4.5.
Geekbench6: 2,570 points in Geekbench 6 single-core and around 23,126 points in Geekbench 6 multi-core tests. https://www.tomshardwar
it is a pretty low pricer to get up to 128GB RAM though compared to all other devices!
If the machine wasn't so god damn ugly looking I might try one.
but then again I didn't need my 128GB Macbook M5 Max so I don't think I'll use up all of the memory in this RTX ARM laptop either. I wonder if the software suite still sucks ass for Windows ARM. lol. (for me - I'm sure many will say it works just fine but those people so far never used snapdragon + windows arm-chair qb)
I thought someone said Linux may actually get support since Nvidia is behind this chip instead Snapdragon -- which appears to be dying even slower death given RTX Spark cpu is out.
Join The Conversation
Share your experience with the Slickdeals community
Share information with the community. Please follow our Community Guidelines and be kind!