PantheonSim / GPUs / NVIDIA RTX PRO 6000 Blackwell Workstation Edition

Simulated NVIDIA RTX PRO 6000 Blackwell Workstation Edition: a GPU profile for CUDA and ROCm testing

PantheonSim, the GPU simulator for CUDA and ROCm, simulates the NVIDIA RTX PRO 6000 Blackwell Workstation Edition (Blackwell, 12.0) as the profile nvidia/rtx-pro-6000. The Workstation Edition of the RTX PRO 6000 Blackwell, a 96 GB GDDR7 card on the same GB202 die as the Server Edition. Its profile is built from public documents and not yet measured against a physical card.

The profile at a glance

These are the numbers a program can ask this simulated GPU, and the values nvidia-smi show for it. They come from the profile file in the simulator's repository, pinned to one commit of it.

Property	Value
Profile id	`nvidia/rtx-pro-6000`
Vendor and architecture	NVIDIA, Blackwell
Compute capability	`12.0`
Device memory	96.0 GiB, GDDR
Streaming multiprocessors	188
Warp size	32
Max threads per multiprocessor	1536
Shared memory per block (opt-in maximum)	99 KiB
L2 cache	128 MiB
Memory bus width	512-bit
Power limit (profile)	600 W
Maximum clocks (profile)	2617 MHz shader, 14000 MHz memory
ECC memory	yes
PCIe link reported	Gen5 x16
Tensor or matrix cores	yes
Verified (measured on a physical card)	not yet measured

What this GPU is for

The Workstation Edition of the RTX PRO 6000 Blackwell, a 96 GB GDDR7 card on the same GB202 die as the Server Edition.

Blackwell is split across several compute capabilities in these profiles: 10.0 for the B200 and GB200, 10.3 for the B300, 11.0 for Jetson Thor, 12.0 for the GeForce RTX 50 series and the RTX PRO 6000 cards, and 12.1 for the GB10. Only the RTX PRO 6000 Server Edition was read from a physical card.

Among the 37 NVIDIA profiles it ranks 9th by device memory (96.0 GiB) and 2nd by SMs (188), counting the largest as first.

Is this profile verified?

No. This profile is marked not verified: it is built from public documents only, and nothing in it was measured on a physical card. Some figures are derived or copied from the nearest profile where the vendor publishes none, so treat them as approximate, and memory sizes as upper bounds. Programs still run on it, and it still answers every call as the architecture's rules require, but a value such as a clock or a limit may differ from a real card's.

Simulate this GPU

Build the simulator (see the docs) or open the playground, then:

nvcc -cudart shared -arch=compute_120 my_app.cu -o my_app
vgpu run --gpu nvidia/rtx-pro-6000 ./my_app
vgpu shell -y --gpu nvidia/rtx-pro-6000 --count 4
vgpu info --gpu nvidia/rtx-pro-6000

This profile is compute capability 12.0. A kernel built for compute_75 runs on it and on every other NVIDIA profile; a kernel built for a newer architecture is refused by a profile that is older than it. Blackwell targets need CUDA 12.8 or newer. Link the CUDA runtime shared (-cudart shared) so the simulator can stand in for it.

To use it every time, put gpu: nvidia/rtx-pro-6000 in a pantheonsim.yaml settings file, or in a GitHub Actions job give the PantheonSim action gpu: nvidia/rtx-pro-6000 (see CUDA and HIP tests in GitHub Actions). The simulator checks behaviour and has no timing model, so a run on a physical GPU before a release still counts. The CUDA guide explains the rest.

Where the numbers come from

Each profile file begins with a note on its sources. This is the note for NVIDIA RTX PRO 6000 Blackwell Workstation Edition, as written in the file:

Source note for nvidia/rtx-pro-6000

(GB202, sm_120). Public values; verified: false, nothing here has been measured on a card. Where each value comes from: - compute capability 12.0: NVIDIA's CUDA GPUs table (developer.nvidia.com/cuda-gpus). - SMs 188: NVIDIA RTX Blackwell GPU Architecture whitepaper (v1.0), Table 1 (RTX PRO 6000 Blackwell Workstation Edition: 188 SMs, 24064 CUDA Cores at 128 an SM, 752 Tensor Cores); the full GB202 has 192. The Max-Q data sheet gives the same 24,064 cores. - memory 96 GB GDDR7 with ECC, 512-bit, 28 Gbps, 1792 GB/s, boost clock 2617 MHz, 600 W Total Graphics Power, PCIe Gen 5: the same table and NVIDIA's product page. NVIDIA's "GB" for graphics memory is binary (see rtx2080ti.yaml), so vram_bytes is 98304 MiB; a real card's nvidia-smi total is a little smaller, so this is an upper bound. - PCI device 0x2BB1, "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", from the name table in NVIDIA's open GPU kernel modules (g_nv_name_released.h). - Per-SM limits are compute capability 12.0's, the same as nvidia/rtx5090: the CUDA C++ Programming Guide's table (64K registers, 1536 threads, 24 blocks, 100 KB shared memory an SM, 99 KB a block), and the whitepaper's 256 KB register file an SM. - CUDA attributes (the `cuda:` section and l2_cache_bytes): L2 131072 KB and the 512-bit bus: RTX Blackwell whitepaper; persisting L2 is 3/4 of the L2, nvidia/rtx5090's ratio.

The GPU index lists every profile, and this one is not yet measured.

Other Blackwell profiles: