I'd love to hear what you think! Please drop me a line and let me know what you like and what could be better. 🙏
NVIDIA RTX A5000 24GB Specs, Benchmarks & Pricing
The NVIDIA RTX A5000 GPU is a high-end graphics card designed for demanding applications, including machine learning and advanced rendering. It features NVIDIA's GA102 GPU architecture and offers a substantial 24 GB of GDDR6 memory. The RTX A5000 is known for its balance between performance, power consumption, and memory capabilities, making it a versatile choice for a range of AI and ML workloads.
- Release Date: April 12, 2021
- MSRP: $2,250 USD
- GPU Architecture: Ampere
- Hardware-Accelerated GEMM Operations:FP64 FP32 TF32 BF16 FP16 FP8 FP4 INT8 INT4
- CUDA Compute Capability : 8.6
Strengths
- 24GB VRAM for model and data capacity
- 27.77 TFLOPS FP32 compute performance
Specifications for NVIDIA RTX A5000
| Specification | Performance Ranking |
|---|---|
| FP32 TFLOPs | |
| FP16 TFLOPs | |
| Tensor Core Count | |
| Memory Capacity (GB) | |
| Memory Bandwidth (GB/s) | |
| Int8 TOPs | |
| FP8 TFLOPs | Data not available |
| FP4 TFLOPs | Data not available |
Real-time NVIDIA RTX A5000 GPU Prices
We're tracking 66 of the NVIDIA RTX A5000 GPUs currently available for sale. The lowest average price is $2,296 (what's this?)
Buy NowCompare Price/Performance to other GPUs
We track real-time prices of other GPUs too so that you can compare the price/performance of the NVIDIA RTX A5000 GPU to other GPUs.
Compare GPU Price/PerformanceCompare NVIDIA RTX A5000 to Another GPU
Compare the NVIDIA RTX A5000 directly to another GPU to see specs, benchmarks, and prices side-by-side.
Compare GPUs Side-by-SidePrice History
NVIDIA RTX A5000 Price History
This chart tracks NVIDIA RTX A5000 prices over the past 6 months based on eBay listings. The yellow line shows the average of the three lowest-priced listings each day—a useful indicator for finding deals.
Trend: Over this period, the lowest average price has been up 27.6%. The current lowest average is $2,296, compared to a typical $1,925 over the period. The lowest average price is calculated as the average of the three lowest-priced listings each day.
gpupoet.com
Product Identifiers
Manufacturer Part Numbers (6)
- Board ID
- PG132
- Product SKU
- PG132 SKU 500
- NVIDIA Part Number
- 900-5G132-2200-000
- NVIDIA Part Number
- 900-5G132-2500-000
- NVIDIA Part Number
- 900-5G132-1700-000
- NVIDIA Part Number
- 900-5G132-0100-001
Available from 6 Partners (7 products)
PNY
- NVIDIA RTX A5000 24GB GDDR6 Graphics Card
- VCNRTXA5000-PB(sku)
Dell
- NVIDIA RTX A5000 24 GB GDDR6 Graphics Card
- 490-BHTI(part number)
- NVIDIA RTX A5000 24GB GDDR6 Graphics Card
- J4KCH(part number)
HP
- NVIDIA RTX A5000 24GB 4DP GFX
- 20X23AA(part number)
Lenovo
- NVIDIA RTX A5000 24GB GDDR6 Graphics Card
- 4X61D97085(part number)
Supermicro
- NVIDIA RTX A5000 Graphic Card 24 GB GDDR6 ECC
- GPU-NVQRTX-A5000(part number)
ASUS
- NVIDIA RTX A5000 24GB 230W
- 90SKC000-M5LAN0(part number)
References
- https://www.techpowerup.com/gpu-specs/rtx-a5000.c3748
- https://www.nvidia.com/en-us/design-visualization/rtx-a5000/
- https://develop3d.com/hardware/nvidia-rtx-a4000-and-rtx-a5000-gpus-launch/
- https://www.cgchannel.com/2021/04/nvidia-announces-eight-new-professional-ampere-gpus/
- https://forums.developer.nvidia.com/t/looking-for-full-specs-on-nvidia-a5000/217948
- https://www.aspsys.com/nvidia-gpus/
- https://www.nvidia.com/content/PDF/nvidia-ampere-ga-102-gpu-architecture-whitepaper-v2.pdf
- https://www.nvidia.com/en-us/data-center/tensor-cores/
- https://www.serversupply.com/GPU/GDDR6/24GB/NVIDIA/900-5G132-2200-000_399632.htm
- https://pcserverandparts.com/dell-nvidia-rtx-a5000-24gb-gddr6-graphics-card-used-j4kch-900-5g132-0100-001/
- https://www.compeve.com/pciexpress-c-3_32/pny-nvidia-rtx-a5000-vcnrtxa5000-pg132-24gb-gddr6-ecc-4xdp-9005g1321700000-p-15187.html
- https://www.dell.com/en-us/shop/dell-nvidia-rtx-a5000-24-gb-gddr6-full-height-pcie-40x16-4-dp-graphics-card/apd/490-bhti/graphic-video-cards
- https://www.connection.com/product/hp-nvidia-rtx-a5000-24gb-4dp-gfx/20x23aa/41281319
- https://www.connection.com/product/lenovo-nvidia-rtx-a5000-pcie-4.0-x16-graphics-card-24gb-gddr6/4x61d97085/41295516
- https://www.exxactcorp.com/Supermicro-GPU-NVQRTX-A5000-E3770660
- https://smicro.eu/asus-nvidia-rtx-a5000-with-24gb-230w-90skc000-m5lan0-1
- https://forums.developer.nvidia.com/t/what-is-the-fp64-rpeak-value-for-a5000/196056
Notes
- int8TOPS of 222.2 represents non-sparse (dense) INT8 performance, derived from sparse INT8 of 444.4 TOPS ÷ 2. For Ampere architecture, INT8 sparse = 2× FP16 sparse tensor performance (222.2 TFLOPS). Dense value used for consistent cross-vendor comparison. Source: NVIDIA Ampere professional graphics line card and forums.developer.nvidia.com.
- FP16 TFLOPS of 222.2 represents Peak Sparse Tensor Core performance (sourced from NVIDIA datasheet Tensor Performance).
- Note: Previous versions incorrectly listed dense performance (111.1 TFLOPS) as sparse. Updated to match Workstation and Ada generation metrics (222.2 TFLOPS).
- MSRP of $2,250 USD verified from 3 sources - DEVELOP3D notes expected street price of $2,250, CG Channel confirms pricing of $1,000 for RTX A4000 and $2,250 for RTX A5000, and TechPowerUp GPU Database lists launch price.
- Manufacturer identifiers: Board ID PG132 with SKU 500 confirmed from TechPowerUp and multiple reseller listings. Multiple NVPN (NVIDIA Part Numbers) exist for different variants: 900-5G132-2200-000 (standard), 900-5G132-2500-000 (retail), 900-5G132-1700-000 (PNY), and 900-5G132-0100-001 (Dell OEM).
- Third-party products: Part numbers sourced from official vendor websites and authorized resellers. PNY uses VCNRTXA5000-PB as their primary SKU. Dell offers two part numbers (490-BHTI and J4KCH/0J4KCH) for different configurations. HP uses 20X23AA. Lenovo uses 4X61D97085. Supermicro uses GPU-NVQRTX-A5000. ASUS uses 90SKC000-M5LAN0.
- Phase 5 audit (2026-09-02): gpuArchitecture normalized from lowercase 'ampere' to canonical 'Ampere' capitalization per the canonical table in .claude/agents/gpu-researcher.md.
- Phase 5 audit (2026-09-02): FP64 removed from supportedHardwareOperations. The RTX A5000 uses the GA102 die (confirmed in the summary above: "NVIDIA's GA102 GPU architecture"), not GA100. Per an NVIDIA staff reply on this exact GPU's NVIDIA Developer Forums thread, 'The Tensor core units in the GA102 chip architecture do not contain the FP64 path that is present in e.g. the Tensor core units in the A100 GPU' (https://forums.developer.nvidia.com/t/what-is-the-fp64-rpeak-value-for-a5000/196056). Canonical Ampere baseline (TF32, BF16, FP16, INT8, INT4) applied per .claude/agents/gpu-researcher.md; FP64 Tensor Core support is reserved for GA100-based SKUs (A100, A30) only.
- fp8TFLOPS and fp4TFLOPS are left null (field omitted) and not added to unverifiedSpecs — this is a confirmed hardware negative, not a research gap. The RTX A5000's Ampere 3rd-generation Tensor Cores predate both FP8 (introduced in Ada Lovelace/Hopper) and FP4 (introduced in Blackwell) tensor formats, per the canonical architecture table in .claude/agents/gpu-researcher.md.