Radeon AI PRO R9700 vs GeForce RTX 3090 Ti
Aggregate performance score
We've compared GeForce RTX 3090 Ti with Radeon AI PRO R9700, including specs and performance data.
RTX 3090 Ti outperforms AI PRO R9700 by a small 7% based on our aggregate benchmark results.
Contents
Primary details
GPU architecture, market segment, value for money and other general parameters compared.
| Place in the ranking | 19 | 29 |
| Place by popularity | not in top-100 | not in top-100 |
| Cost-effectiveness evaluation | 17.59 | 28.31 |
| Power efficiency | 12.08 | 16.93 |
| Architecture | Ampere (2020−2025) | RDNA 4.0 (2025−2026) |
| GPU code name | GA102 | Navi 48 |
| Market segment | Desktop | Workstation |
| Release date | 27 January 2022 (4 years ago) | 23 July 2025 (1 year ago) |
| Launch price (MSRP) | $1,999 | $1,299 |
Cost-effectiveness evaluation
The higher the ratio, the better. We use the manufacturer's recommended prices.
AI PRO R9700 has 61% better value for money than RTX 3090 Ti.
Performance to price scatter graph
Currently popular graphics cards are shown for comparison.
Detailed specifications
General parameters such as number of shaders, GPU core base clock and boost clock speeds, manufacturing process, texturing and calculation speed. Real power consumption of some graphics cards can well exceed their nominal TDP, especially if overclocked.
| Pipelines / CUDA cores | 10752 | 4096 |
| Core clock speed | 1560 MHz | 1660 MHz |
| Boost clock speed | 1860 MHz | 2920 MHz |
| Number of transistors | 28,300 million | 53,900 million |
| Manufacturing process technology | 8 nm | 4 nm |
| Power consumption (TDP) | 450 Watt | 300 Watt |
| Texture fill rate | 625.0 | 747.5 |
| Floating-point processing power | 40 TFLOPS | 47.84 TFLOPS |
| ROPs | 112 | 128 |
| TMUs | 336 | 256 |
| Tensor Cores | 336 | 128 |
| Ray Tracing Cores | 84 | 64 |
| L0 Cache | no data | 1 MB |
| L1 Cache | 10.5 MB | no data |
| L2 Cache | 6 MB | 8 MB |
| L3 Cache | no data | 64 MB |
Form factor & compatibility
Information on compatibility with other computer components. Useful when choosing a future computer configuration or upgrading an existing one. For desktop graphics cards it's interface and bus (motherboard compatibility), additional power connectors (power supply compatibility).
| Interface | PCIe 4.0 x16 | PCIe 5.0 x16 |
| Length | 336 mm | 267 mm |
| Width | 3-slot | 2-slot |
| Supplementary power connectors | 1x 16-pin | 1x 16-pin |
VRAM capacity and type
Parameters of VRAM installed: its type, size, bus, clock and resulting bandwidth. Integrated GPUs have no dedicated video RAM and use a shared part of system RAM.
| Memory type | GDDR6X | GDDR6 |
| Maximum RAM amount | 24 GB | 32 GB |
| Memory bus width | 384 Bit | 256 Bit |
| Memory clock speed | 1313 MHz | 2518 MHz |
| Memory bandwidth | 1.01 TB/s | 644.6 GB/s |
| Shared memory | - | - |
| Resizable BAR | + | + |
Connectivity and outputs
This section shows the types and number of video connectors on each GPU. The data applies specifically to desktop reference models (for example, NVIDIA’s Founders Edition). OEM partners often modify both the number and types of ports. On notebook GPUs, video‐output options are determined by the laptop’s design rather than the graphics chip itself.
| Display Connectors | 1x HDMI 2.1, 3x DisplayPort 1.4a | 1x HDMI 2.1b, 3x DisplayPort 2.1a |
| HDMI | + | + |
API and SDK support
List of supported 3D and general-purpose computing APIs, including their specific versions.
| DirectX | 12 Ultimate (12_2) | 12 Ultimate (12_2) |
| Shader Model | 6.8 | 6.8 |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 2.2 |
| Vulkan | 1.4 | 1.3 |
| CUDA | 8.6 | - |
| DLSS | + | + |
Synthetic benchmarks
Non-gaming benchmark results comparison. The combined score is measured on a 0-100 point scale.
Combined synthetic benchmark score
This is our combined benchmark score.
Passmark
This is the most ubiquitous GPU benchmark. It gives the graphics card a thorough evaluation under various types of load, providing four separate benchmarks for Direct3D versions 9, 10, 11 and 12 (the last being done in 4K resolution if possible), and few more tests engaging DirectCompute capabilities.
Gaming performance
Let's see how good the compared graphics cards are for gaming. Particular gaming benchmark results are measured in FPS.
Average FPS across all PC games
Here are the average frames per second in a large set of popular games across different resolutions:
| Full HD | 210
+10.5%
| 190−200 |
| 1440p | 142
+9.2%
| 130−140 |
| 4K | 101 | 90−95
−10.9%
|
Cost per frame, $
| 1080p | 9.52 | 6.84
+28.2%
|
| 1440p | 14.08 | 9.99
+29%
|
| 4K | 19.79 | 14.43
+27.1%
|
- AI PRO R9700 has 28% lower cost per frame in 1080p
- AI PRO R9700 has 29% lower cost per frame in 1440p
- AI PRO R9700 has 27% lower cost per frame in 4K
FPS performance in popular games
Full HD
Low
| Atomic Heart | 210−220
+8%
|
200−210 |
| Counter-Strike 2 | 300−350
+10.7%
|
290−300 |
| Cyberpunk 2077 | 219
+9.5%
|
200−210 |
Full HD
Medium
| Atomic Heart | 210−220
+8%
|
200−210 |
| Battlefield 5 | 180−190
+7.6%
|
170−180 |
| Counter-Strike 2 | 300−350
+10.7%
|
290−300 |
| Cyberpunk 2077 | 201
+11.7%
|
180−190 |
| Far Cry 5 | 200−210
+7.9%
|
190−200 |
| Fortnite | 300−350
+7.9%
|
280−290 |
| Forza Horizon 4 | 280−290
+8.8%
|
260−270 |
| PLAYERUNKNOWN'S BATTLEGROUNDS | 170−180
+8.1%
|
160−170 |
| Valorant | 400−450
+22%
|
350−400 |
Full HD
High
| Atomic Heart | 210−220
+8%
|
200−210 |
| Battlefield 5 | 180−190
+7.6%
|
170−180 |
| Counter-Strike 2 | 300−350
+10.7%
|
290−300 |
| Counter-Strike: Global Offensive | 270−280
+11.2%
|
250−260 |
| Cyberpunk 2077 | 173
+8.1%
|
160−170 |
| Dota 2 | 217
+8.5%
|
200−210 |
| Far Cry 5 | 200−210
+7.9%
|
190−200 |
| Far Cry 6 | 161
+7.3%
|
150−160 |
| Fortnite | 300−350
+7.9%
|
280−290 |
| Forza Horizon 4 | 280−290
+8.8%
|
260−270 |
| Grand Theft Auto V | 170
+13.3%
|
150−160 |
| Metro Exodus | 178
+11.3%
|
160−170 |
| PLAYERUNKNOWN'S BATTLEGROUNDS | 170−180
+8.1%
|
160−170 |
| The Witcher 3: Wild Hunt | 394
+12.6%
|
350−400 |
| Valorant | 400−450
+22%
|
350−400 |
Full HD
Ultra
| Battlefield 5 | 180−190
+7.6%
|
170−180 |
| Cyberpunk 2077 | 152
+8.6%
|
140−150 |
| Dota 2 | 195
+8.3%
|
180−190 |
| Far Cry 5 | 200−210
+7.9%
|
190−200 |
| Forza Horizon 4 | 280−290
+8.8%
|
260−270 |
| PLAYERUNKNOWN'S BATTLEGROUNDS | 170−180
+8.1%
|
160−170 |
| The Witcher 3: Wild Hunt | 193
+7.2%
|
180−190 |
| Valorant | 400−450
+22%
|
350−400 |
Full HD
Epic
| Fortnite | 300−350
+7.9%
|
280−290 |
1440p
High
| Counter-Strike 2 | 220−230
+11%
|
200−210 |
| Counter-Strike: Global Offensive | 500−550
+14.7%
|
450−500 |
| Far Cry 6 | 133
+10.8%
|
120−130 |
| Grand Theft Auto V | 151
+7.9%
|
140−150 |
| Metro Exodus | 125
+13.6%
|
110−120 |
| PLAYERUNKNOWN'S BATTLEGROUNDS | 170−180
+9.4%
|
160−170 |
| Valorant | 450−500
+7.8%
|
450−500 |
1440p
Ultra
| Battlefield 5 | 180−190
+8.8%
|
170−180 |
| Cyberpunk 2077 | 104
+9.5%
|
95−100 |
| Far Cry 5 | 180−190
+7.6%
|
170−180 |
| Forza Horizon 4 | 240−250
+7.8%
|
230−240 |
| The Witcher 3: Wild Hunt | 180−190
+12.5%
|
160−170 |
1440p
Epic
| Fortnite | 150−160
+7.9%
|
140−150 |
4K
High
| Atomic Heart | 65−70
+13.3%
|
60−65 |
| Counter-Strike 2 | 100−105
+11.1%
|
90−95 |
| Far Cry 6 | 90
+12.5%
|
80−85 |
| Grand Theft Auto V | 181
+13.1%
|
160−170 |
| Metro Exodus | 84
+12%
|
75−80 |
| The Witcher 3: Wild Hunt | 173
+8.1%
|
160−170 |
| Valorant | 300−350
+9.7%
|
290−300 |
4K
Ultra
| Battlefield 5 | 130−140
+13.3%
|
120−130 |
| Counter-Strike 2 | 100−105
+11.1%
|
90−95 |
| Cyberpunk 2077 | 53
+17.8%
|
45−50 |
| Dota 2 | 184
+8.2%
|
170−180 |
| Far Cry 5 | 120−130
+10%
|
110−120 |
| Forza Horizon 4 | 200−210
+7.4%
|
190−200 |
| PLAYERUNKNOWN'S BATTLEGROUNDS | 95−100
+12.9%
|
85−90 |
4K
Epic
| Fortnite | 75−80
+12.9%
|
70−75 |
This is how RTX 3090 Ti and AI PRO R9700 compete in popular games:
- RTX 3090 Ti is 11% faster in 1080p
- RTX 3090 Ti is 9% faster in 1440p
- RTX 3090 Ti is 12% faster in 4K
Pros & cons summary
| Performance score | 69.80 | 65.20 |
| Recency | 27 January 2022 | 23 July 2025 |
| Maximum RAM amount | 24 GB | 32 GB |
| Chip lithography | 8 nm | 4 nm |
| Power consumption (TDP) | 450 Watt | 300 Watt |
RTX 3090 Ti has a 7% higher aggregate performance score.
AI PRO R9700, on the other hand, has an age advantage of 3 years, a 33% higher maximum VRAM amount, a 50% more advanced lithography process, and 33% lower power consumption.
Given the minimal performance differences, no clear winner can be declared between GeForce RTX 3090 Ti and Radeon AI PRO R9700.
Be aware that GeForce RTX 3090 Ti is a desktop graphics card while Radeon AI PRO R9700 is a workstation one.
Other comparisons
We selected several comparisons of graphics cards with performance close to those reviewed, providing you with more options to consider.
