# Accelerated Computing **URL:** https://forums.developer.nvidia.com/c/accelerated-computing/5.md [Latest](https://forums.developer.nvidia.com/latest.md) · [Categories](https://forums.developer.nvidia.com/categories.md) · [Tags](https://forums.developer.nvidia.com/tags.md) --- ## [Unable to get 7.2x working on DGX spark with cubb 26-1 and Benetel O-RU](https://forums.developer.nvidia.com/t/unable-to-get-7-2x-working-on-dgx-spark-with-cubb-26-1-and-benetel-o-ru/387045)
**Author:** [@mrunaliniptn](https://forums.developer.nvidia.com/u/mrunaliniptn)\ **Replies:** 4\ **Last updated:** [October 27, 2026, 9:14am UTC](https://forums.developer.nvidia.com/t/unable-to-get-7-2x-working-on-dgx-spark-with-cubb-26-1-and-benetel-o-ru/387045 "2026-10-27T09:14:00Z")
Hi, I installed Aerial L1(26.1), OAI L2/L3(ATB1.1) on DGX Spark and using Benetel as O-RU(v1.0.4). PTP is working fine on both sides. Whenever the UE sends the RRCSetupComplete message(MSG5), "SCF\_ERROR\_CODE\_L1\_MISSING\_… --- ## [Native vLLM for RTX Spark on Windows ARM64 (w/Qwen3.8-Flash-Next-FP4-Local)](https://forums.developer.nvidia.com/t/native-vllm-for-rtx-spark-on-windows-arm64-w-qwen3-8-flash-next-fp4-local/387069)
**Author:** [@dev.parkjeonghwan](https://forums.developer.nvidia.com/u/dev.parkjeonghwan)\ **Replies:** 0\ **Last updated:** [October 27, 2026, 8:56am UTC](https://forums.developer.nvidia.com/t/native-vllm-for-rtx-spark-on-windows-arm64-w-qwen3-8-flash-next-fp4-local/387069 "2026-10-27T08:56:05Z")
Hi everyone, I’ve released a native Windows ARM64 port of vLLM for RTX Spark N1X. No Linux or WSL is needed. It is based on vLLM 0.32.0 and myllmbox’s v5.3.1 recipe, and currently supports NVIDIA’s Qwen3.8-Flash-Next-F… --- ## [New Deepseek 4.1 flash recipe](https://forums.developer.nvidia.com/t/new-deepseek-4-1-flash-recipe/384438)
**Author:** [@christopher\_owen](https://forums.developer.nvidia.com/u/christopher_owen)\ **Replies:** 73\ **Last updated:** [October 27, 2026, 8:47am UTC](https://forums.developer.nvidia.com/t/new-deepseek-4-1-flash-recipe/384438 "2026-10-27T08:47:08Z")
Here’s my version, if anyone is interested. It’s tuned to my workload and setup on a 3x switchless FE setup. --- ## [DeepSeek-V4.1-Flash-1017 (EXL3 2.9bpw) on 2x DGX Spark with TensorFold: 57 tok/s text, 92 tok/s code, 2.1-2.3x the vLLM kit](https://forums.developer.nvidia.com/t/deepseek-v4-1-flash-1017-exl3-2-9bpw-on-2x-dgx-spark-with-tensorfold-57-tok-s-text-92-tok-s-code-2-1-2-3x-the-vllm-kit/386125)
**Author:** [@Elevatormusic](https://forums.developer.nvidia.com/u/Elevatormusic)\ **Replies:** 247\ **Last updated:** [October 27, 2026, 8:23am UTC](https://forums.developer.nvidia.com/t/deepseek-v4-1-flash-1017-exl3-2-9bpw-on-2x-dgx-spark-with-tensorfold-57-tok-s-text-92-tok-s-code-2-1-2-3x-the-vllm-kit/386125 "2026-10-27T08:23:45Z")
DeepSeek-V4.1-Flash-1017 (EXL3 2.9 bpw, 1M context) on 2x DGX Spark, TP=2 over the CX7 cable. Recipe Github: GitHub - Elevatormusic/DeepSeek-V4.1-Flash-1017-EXL3-TensorFold-2x-DGX-Spark: DeepSeek-V4.1-Flash-1017 EXL3 2… --- ## [Reclaim 2GiB of ram on headless sparks](https://forums.developer.nvidia.com/t/reclaim-2gib-of-ram-on-headless-sparks/384621)
**Author:** [@jontaylor](https://forums.developer.nvidia.com/u/jontaylor)\ **Replies:** 23\ **Last updated:** [October 27, 2026, 8:19am UTC](https://forums.developer.nvidia.com/t/reclaim-2gib-of-ram-on-headless-sparks/384621 "2026-10-27T08:19:28Z")
Thanks to @emihuang for discovering that the GB10’s display-reserved memory could be used with CUDA—that sparked a slightly different experiment! With more than a little help from Codex, I’ve managed to reclaim 2,046 Mi… --- ## [Aerial 26-2 / ARC Pro / 4T4R 11-cell: test case and TV configuration](https://forums.developer.nvidia.com/t/aerial-26-2-arc-pro-4t4r-11-cell-test-case-and-tv-configuration/387059)
**Author:** [@august.lai](https://forums.developer.nvidia.com/u/august.lai)\ **Replies:** 0\ **Last updated:** [October 27, 2026, 7:31am UTC](https://forums.developer.nvidia.com/t/aerial-26-2-arc-pro-4t4r-11-cell-test-case-and-tv-configuration/387059 "2026-10-27T07:31:57Z")
Hi @nhashimoto, We would like to test the Aerial 26-2 MGX ARC Pro reported in the release notes: 11 peak 4T4R cells 4 DL layers and 2 UL layers per UE per cell 1.46 Gbps DL and 210 Mbps UL per cell 400G fron… --- ## [Let’s optimize DeepSeek-V4.1-Flash-1017 for 2× DGX Spark](https://forums.developer.nvidia.com/t/let-s-optimize-deepseek-v4-1-flash-1017-for-2x-dgx-spark/386574)
**Author:** [@PILCOTHINK](https://forums.developer.nvidia.com/u/PILCOTHINK)\ **Replies:** 4\ **Last updated:** [October 27, 2026, 6:41am UTC](https://forums.developer.nvidia.com/t/let-s-optimize-deepseek-v4-1-flash-1017-for-2x-dgx-spark/386574 "2026-10-27T06:41:55Z")
Serving Guide Model Test Result | model | test | t/s | peak t/s | ttfr (ms) | est\_ppt (ms) | e2e\_ttft (ms) | |:--------------------|---------------:|-----------… --- ## [Up to 70tok/s Qwen3.8-Flash-Next-Int4-AutoRound](https://forums.developer.nvidia.com/t/up-to-70tok-s-qwen3-8-flash-next-int4-autoround/382733)
**Author:** [@azampatti](https://forums.developer.nvidia.com/u/azampatti)\ **Replies:** 317\ **Last updated:** [October 27, 2026, 5:49am UTC](https://forums.developer.nvidia.com/t/up-to-70tok-s-qwen3-8-flash-next-int4-autoround/382733 "2026-10-27T05:49:17Z")
Just like I did with my Faster Qwen3.5 122B, I’m basically doing the same with this great awesome new model from Qwen. Single-line command: git clone https://github.com/azampatti/Qwen3.8-Flash-Next-Int4-FAST.git && cd … --- ## [Is pypi.ngc.nvidia.com Python Repository Decommissioned?](https://forums.developer.nvidia.com/t/is-pypi-ngc-nvidia-com-python-repository-decommissioned/387048)
**Author:** [@tonydeguia009](https://forums.developer.nvidia.com/u/tonydeguia009)\ **Replies:** 0\ **Last updated:** [October 27, 2026, 5:43am UTC](https://forums.developer.nvidia.com/t/is-pypi-ngc-nvidia-com-python-repository-decommissioned/387048 "2026-10-27T05:43:29Z")
Hello, Good day. I just would like to check, is the pypi.ngc.nvidia.com actually decommissioned now? It seems that it is not working anymore on DNS level, so it is causing some problems on older container images that ar… --- ## [DGX Spark vs RTX Spark](https://forums.developer.nvidia.com/t/dgx-spark-vs-rtx-spark/372316)
**Author:** [@kenrb42](https://forums.developer.nvidia.com/u/kenrb42)\ **Replies:** 27\ **Last updated:** [October 27, 2026, 5:24am UTC](https://forums.developer.nvidia.com/t/dgx-spark-vs-rtx-spark/372316 "2026-10-27T05:24:22Z")
Pardon me if this has been discussed and answered already, but I am still struggling to find info around it, and perhaps we simply just do not know yet. DGX Spark vs RTX Spark. What is really the differences here? Why a… --- ## [Spark-DSV41: Production-Hardened DeepSeek-V4.1-Flash-1017 Engine & Go Control Plane for 2x DGX Spark](https://forums.developer.nvidia.com/t/spark-dsv41-production-hardened-deepseek-v4-1-flash-1017-engine-go-control-plane-for-2x-dgx-spark/387041)
**Author:** [@massoud.mazar](https://forums.developer.nvidia.com/u/massoud.mazar)\ **Replies:** 0\ **Last updated:** [October 27, 2026, 3:21am UTC](https://forums.developer.nvidia.com/t/spark-dsv41-production-hardened-deepseek-v4-1-flash-1017-engine-go-control-plane-for-2x-dgx-spark/387041 "2026-10-27T03:21:36Z")
GitHub: GitHub - mazar/spark-dsv41: High-Throughput DeepSeek-V4.1-Flash-1017 Engine & Go Control Plane for 2x NVIDIA DGX Spark (sm\_121a) · GitHub Spark-DSV41 is a self-contained distributed inference stack and compiled … --- ## [DGX Monarch - Dual Spark Rendering w/ ComfyUI](https://forums.developer.nvidia.com/t/dgx-monarch-dual-spark-rendering-w-comfyui/385450)
**Author:** [@Balaxxe](https://forums.developer.nvidia.com/u/Balaxxe)\ **Replies:** 17\ **Last updated:** [October 27, 2026, 2:47am UTC](https://forums.developer.nvidia.com/t/dgx-monarch-dual-spark-rendering-w-comfyui/385450 "2026-10-27T02:47:49Z")
For those of you who have been wanting to utilize your dual Spark cluster for image/video gen with ComfyUI - today is your day. I started this personal project back in April of this year, and it’s at a point where I’m f… --- ## [GIGABYTE AI TOP ATOM: stopped fan, severe heat and loss of access after updates](https://forums.developer.nvidia.com/t/gigabyte-ai-top-atom-stopped-fan-severe-heat-and-loss-of-access-after-updates/387012)
**Author:** [@haukebr](https://forums.developer.nvidia.com/u/haukebr)\ **Replies:** 2\ **Last updated:** [October 27, 2026, 1:19am UTC](https://forums.developer.nvidia.com/t/gigabyte-ai-top-atom-stopped-fan-severe-heat-and-loss-of-access-after-updates/387012 "2026-10-27T01:19:35Z")
I have two GIGABYTE AI TOP ATOM (NVIDIA GB10) systems used as a headless, two-node inference cluster. After today’s software updates and testing, the head became unreachable. On physically checking the machines, I found … --- ## [Stuck GPU clock on GB10 (500–800 MHz, zero throttle reasons): a measured case on a Gigabyte AI TOP ATOM, a detector, and a heat side-effect](https://forums.developer.nvidia.com/t/stuck-gpu-clock-on-gb10-500-800-mhz-zero-throttle-reasons-a-measured-case-on-a-gigabyte-ai-top-atom-a-detector-and-a-heat-side-effect/386060)
**Author:** [@lcasarin](https://forums.developer.nvidia.com/u/lcasarin)\ **Replies:** 4\ **Last updated:** [October 27, 2026, 1:08am UTC](https://forums.developer.nvidia.com/t/stuck-gpu-clock-on-gb10-500-800-mhz-zero-throttle-reasons-a-measured-case-on-a-gigabyte-ai-top-atom-a-detector-and-a-heat-side-effect/386060 "2026-10-27T01:08:10Z")
This is another confirmed case of the state described in threads 376039, 376239, 366944, 362737, 375687, 376542 and 378150, with numbers and a detector. Unit: Gigabyte AI TOP ATOM (GB10), DGX OS 7.2.3, driver 580.178.04… --- ## [DGX Spark: log-less hard power-off + implausible ACPI thermal sensor spikes — Field Diagnostics PASS (case 261014-000313)](https://forums.developer.nvidia.com/t/dgx-spark-log-less-hard-power-off-implausible-acpi-thermal-sensor-spikes-field-diagnostics-pass-case-261014-000313/386556)
**Author:** [@marioreirizf](https://forums.developer.nvidia.com/u/marioreirizf)\ **Replies:** 1\ **Last updated:** [October 27, 2026, 1:07am UTC](https://forums.developer.nvidia.com/t/dgx-spark-log-less-hard-power-off-implausible-acpi-thermal-sensor-spikes-field-diagnostics-pass-case-261014-000313/386556 "2026-10-27T01:07:25Z")
Hi, NVIDIA Customer Care (case 261014-000313, Jason @ NVCC/L2) asked me to post here so DGX staff can assess whether an RMA is needed. summary-fieldiag-2026-10-22.txt (3.0 KB) Unit: DGX Spark FE (serial in the case). D… --- ## [Running Kokoro TTS on NVIDIA DGX Spark (ARM64/GB10)](https://forums.developer.nvidia.com/t/running-kokoro-tts-on-nvidia-dgx-spark-arm64-gb10/368846)
**Author:** [@swann.schilling](https://forums.developer.nvidia.com/u/swann.schilling)\ **Replies:** 1\ **Last updated:** [October 27, 2026, 1:00am UTC](https://forums.developer.nvidia.com/t/running-kokoro-tts-on-nvidia-dgx-spark-arm64-gb10/368846 "2026-10-27T01:00:16Z")
I successfully got Kokoro TTS running on the DGX Spark with full GPU acceleration. Here’s a step-by-step guide for anyone else looking to deploy this. Also Japanese voices ready (which was the biggest painpoint). What i… --- ## [RE::-Overheating and failing the Nvidia Fieldtest Diag](https://forums.developer.nvidia.com/t/re-overheating-and-failing-the-nvidia-fieldtest-diag/387019)
**Author:** [@pistonspy.mobile](https://forums.developer.nvidia.com/u/pistonspy.mobile)\ **Replies:** 1\ **Last updated:** [October 27, 2026, 12:59am UTC](https://forums.developer.nvidia.com/t/re-overheating-and-failing-the-nvidia-fieldtest-diag/387019 "2026-10-27T00:59:24Z")
I am on my second unit and worried about the short 12 month warranty. Does NVidia accept this as a manufacturing defect and will honour this post warranty or could we be stuck with a really expensive paper weight ? It … --- ## [NVIDIA official Qwen3.8 Flash Next checkpoint for single DGX Spark](https://forums.developer.nvidia.com/t/nvidia-official-qwen3-8-flash-next-checkpoint-for-single-dgx-spark/385994)
**Author:** [@luisiam](https://forums.developer.nvidia.com/u/luisiam)\ **Replies:** 35\ **Last updated:** [October 27, 2026, 12:54am UTC](https://forums.developer.nvidia.com/t/nvidia-official-qwen3-8-flash-next-checkpoint-for-single-dgx-spark/385994 "2026-10-27T00:54:22Z")
NVIDIA released a checkpoint and recipe to run Qwen3.8 Flash Next on a single DGX Spark. Seems like NVIDIA still remember the spark owner. =\] --- ## [Is your GB10 stuck in low-power mode? A free 45-second check, healthy vs stuck numbers, and the power-drain fix](https://forums.developer.nvidia.com/t/is-your-gb10-stuck-in-low-power-mode-a-free-45-second-check-healthy-vs-stuck-numbers-and-the-power-drain-fix/387029)
**Author:** [@Anantus](https://forums.developer.nvidia.com/u/Anantus)\ **Replies:** 0\ **Last updated:** [October 26, 2026, 11:45pm UTC](https://forums.developer.nvidia.com/t/is-your-gb10-stuck-in-low-power-mode-a-free-45-second-check-healthy-vs-stuck-numbers-and-the-power-drain-fix/387029 "2026-10-26T23:45:58Z")
Two of the eight GB10s in our cluster spent days running at less than half speed with no error, no log entry and no throttle reason. If your Spark (or GX10, or any GB10) suddenly feels slower than everyone else’s numbers… --- ## [How safe is it to update right now?](https://forums.developer.nvidia.com/t/how-safe-is-it-to-update-right-now/386696)
**Author:** [@torfifrans](https://forums.developer.nvidia.com/u/torfifrans)\ **Replies:** 16\ **Last updated:** [October 26, 2026, 11:41pm UTC](https://forums.developer.nvidia.com/t/how-safe-is-it-to-update-right-now/386696 "2026-10-26T23:41:16Z")
There was an issue with an update a while ago. I haven’t updated in a while, and like everyone, don’t want to destabilize the system. I usually use the series of commands reccomended on nvidia’s site, since mine is headl… --- ## [Nvfortran 26.5 -- TERMINATED Signal 11](https://forums.developer.nvidia.com/t/nvfortran-26-5-terminated-signal-11/377541)
**Author:** [@Don\_Kenzakowski](https://forums.developer.nvidia.com/u/Don_Kenzakowski)\ **Replies:** 3\ **Last updated:** [October 26, 2026, 11:11pm UTC](https://forums.developer.nvidia.com/t/nvfortran-26-5-terminated-signal-11/377541 "2026-10-26T23:11:28Z")
Getting compilation error mesage when using -C option for one of my routines. If I remove the -C out of the compiler option list, the code compiles successfully. \[kenzakow@phoenix SendBug\]$ ./Build.sh ftn=mpif90 FFLA… --- ## [Dgx-spark-networking: open-source switchless RoCE collectives, NCCL profiles and fabric tooling for multi-Spark inference](https://forums.developer.nvidia.com/t/dgx-spark-networking-open-source-switchless-roce-collectives-nccl-profiles-and-fabric-tooling-for-multi-spark-inference/385049)
**Author:** [@christopher\_owen](https://forums.developer.nvidia.com/u/christopher_owen)\ **Replies:** 4\ **Last updated:** [October 26, 2026, 11:07pm UTC](https://forums.developer.nvidia.com/t/dgx-spark-networking-open-source-switchless-roce-collectives-nccl-profiles-and-fabric-tooling-for-multi-spark-inference/385049 "2026-10-26T23:07:49Z")
I’m releasing dgx-spark-networking, an Apache-2.0 library for running tensor-parallel inference across several DGX Sparks over their ConnectX-7 ports, with or without a switch. It packages the fabric configuration that h… --- ## [New DGX Spark: fans barely speed up at high temps and can't be adjusted. Critical. Please open fan control](https://forums.developer.nvidia.com/t/new-dgx-spark-fans-barely-speed-up-at-high-temps-and-cant-be-adjusted-critical-please-open-fan-control/386805)
**Author:** [@maxeffort](https://forums.developer.nvidia.com/u/maxeffort)\ **Replies:** 11\ **Last updated:** [October 26, 2026, 10:49pm UTC](https://forums.developer.nvidia.com/t/new-dgx-spark-fans-barely-speed-up-at-high-temps-and-cant-be-adjusted-critical-please-open-fan-control/386805 "2026-10-26T22:49:47Z")
Hi NVIDIA team, I have a brand new NVIDIA DGX Spark (the original NVIDIA Founders Edition, bought October 23, 2026). Under normal AI workloads the fans stay far too slow, and there is no way to change them. What we see … --- ## [TideLM: DeepSeek-V4.1-Flash, native MXFP4 weights on two DGX Sparks](https://forums.developer.nvidia.com/t/tidelm-deepseek-v4-1-flash-native-mxfp4-weights-on-two-dgx-sparks/387014)
**Author:** [@raspy135](https://forums.developer.nvidia.com/u/raspy135)\ **Replies:** 0\ **Last updated:** [October 26, 2026, 10:02pm UTC](https://forums.developer.nvidia.com/t/tidelm-deepseek-v4-1-flash-native-mxfp4-weights-on-two-dgx-sparks/387014 "2026-10-26T22:02:56Z")
I made an LLM engine called TideLM. TideLM serves DeepSeek flash v4.1 with native MXFP4 quant. It keeps the routed experts your traffic actually uses resident (10,000 of 15,360), learns which ones from the router’s dem… --- ## [DGX Spark fan-control kernel module — running on dual Sparks since mid-August](https://forums.developer.nvidia.com/t/dgx-spark-fan-control-kernel-module-running-on-dual-sparks-since-mid-august/382536)
**Author:** [@christopher\_owen](https://forums.developer.nvidia.com/u/christopher_owen)\ **Replies:** 3\ **Last updated:** [October 26, 2026, 9:52pm UTC](https://forums.developer.nvidia.com/t/dgx-spark-fan-control-kernel-module-running-on-dual-sparks-since-mid-august/382536 "2026-10-26T21:52:34Z")
I’ve observed my DGX Spark reaching approximately 105°C and shutting down during long runs. The firmware contains critical = 104800. NVIDIA doesn’t expose user-space fan control, so I wrote a kernel module to control th… --- ## [Memory usage of std::sort(par\_unseq, ...) on growing range with stdpar](https://forums.developer.nvidia.com/t/memory-usage-of-std-sort-par-unseq-on-growing-range-with-stdpar/327467)
**Author:** [@MaximeRambosson](https://forums.developer.nvidia.com/u/MaximeRambosson)\ **Replies:** 4\ **Last updated:** [October 26, 2026, 9:34pm UTC](https://forums.developer.nvidia.com/t/memory-usage-of-std-sort-par-unseq-on-growing-range-with-stdpar/327467 "2026-10-26T21:34:25Z")
Hello NV team, I wanted to use the std::execution::par\_unseq execution policy to sort a std::vector\ using nvc++ -stdpar, but I encountered an issue: as I gradually increase the size of the range being sorted, GP… --- ## [Different behavior of nvfortran with elemental optional arugments](https://forums.developer.nvidia.com/t/different-behavior-of-nvfortran-with-elemental-optional-arugments/342408)
**Author:** [@JorgeG94](https://forums.developer.nvidia.com/u/JorgeG94)\ **Replies:** 2\ **Last updated:** [October 26, 2026, 9:27pm UTC](https://forums.developer.nvidia.com/t/different-behavior-of-nvfortran-with-elemental-optional-arugments/342408 "2026-10-26T21:27:03Z")
The following program compiles and produces “T” as a result for gfortran, ifort, ifx, flang, and lfortran - with nvfortran this produes a segfault. This seems like a bug. I tried nvfortan 25.7, 25.5, 25.1, 24.9. and 24.1… --- ## [NVFortran bug with repeated name of interface in different modules](https://forums.developer.nvidia.com/t/nvfortran-bug-with-repeated-name-of-interface-in-different-modules/361078)
**Author:** [@christopher.hansen](https://forums.developer.nvidia.com/u/christopher.hansen)\ **Replies:** 2\ **Last updated:** [October 26, 2026, 9:22pm UTC](https://forums.developer.nvidia.com/t/nvfortran-bug-with-repeated-name-of-interface-in-different-modules/361078 "2026-10-26T21:22:40Z")
I am seeing an erroneous compiler error with NVFortran when types in different modules have procedure pointers that share the same interface name, despite the fact the interface is not imported/use-d in the other module.… --- ## [CONTIGUOUS copy-in skipped through assumed-shape wrappers — silent data corruption](https://forums.developer.nvidia.com/t/contiguous-copy-in-skipped-through-assumed-shape-wrappers-silent-data-corruption/361323)
**Author:** [@gio.cava10](https://forums.developer.nvidia.com/u/gio.cava10)\ **Replies:** 2\ **Last updated:** [October 26, 2026, 9:19pm UTC](https://forums.developer.nvidia.com/t/contiguous-copy-in-skipped-through-assumed-shape-wrappers-silent-data-corruption/361323 "2026-10-26T21:19:19Z")
nvfortran 26.1: CONTIGUOUS copy-in silently skipped through assumed-shape wrappers nvfortran 26.1 silently skips the required copy-in when a non-contiguous array section passes through assumed-shape (dimension(:,:,:)) wr… --- ## [DGX Spark: Could Windows on ARM Support Be Possible, Like RTX Spark, N1X?](https://forums.developer.nvidia.com/t/dgx-spark-could-windows-on-arm-support-be-possible-like-rtx-spark-n1x/371870)
**Author:** [@PILCOTHINK](https://forums.developer.nvidia.com/u/PILCOTHINK)\ **Replies:** 73\ **Last updated:** [October 26, 2026, 9:19pm UTC](https://forums.developer.nvidia.com/t/dgx-spark-could-windows-on-arm-support-be-possible-like-rtx-spark-n1x/371870 "2026-10-26T21:19:17Z")
Hello, Recently, I feel like I’ve been seeing a lot of articles about the N1/N1X series. What particularly stood out to me is that they seem to support Windows on ARM. I think I read somewhere that the N1 series and DG… [Next page](https://forums.developer.nvidia.com/c/accelerated-computing/5.md?page=1)