NVIDIA GeForce RTX 5070
The NVIDIA GeForce RTX 5070 is a graphics card released on March 4, 2025, built on the Blackwell architecture using a 4 nm process. It has 6,144 shading units, 12 GB of GDDR7 memory on a 192-bit bus, 672.0 GB/s of memory bandwidth, a boost clock of 2.51 GHz and a 250 W board power rating. It scores 28,677 in the PassMark G3D benchmark. It supports DirectX 12 Ultimate, OpenGL 4.6, Vulkan 1.4 and CUDA compute capability 12.0.
Depth: 62 recorded fields, 24 features, 5 PassMark workloads
Specification data last verified October 1, 2026 against PassMark Videocard Benchmark, NVIDIA GeForce Specifications, NanoReview (Firecrawl extract), NVIDIA Developer: CUDA GPUs, NVIDIA Developer: Video Encode and Decode GPU Support Matrix and Wikipedia GPU list tables (wikigrid-2026-09). Coverage: 10 of 10 tracked specifications recorded.
Performance
The NVIDIA GeForce RTX 5070 scores 28,677 in the PassMark G3D benchmark, higher than 99.1% of the 2,658 GPUs benchmarked (PassMark rank #27), measured across 23,321 submitted results. Its PassMark G2D score is 1,293.
PassMark workload tests
In the PassMark workload tests the NVIDIA GeForce RTX 5070 averages DirectX 9 322 fps, DirectX 10 180 fps, DirectX 11 279 fps, DirectX 12 106 fps and GPU Compute 14,326 ops/s, across 23,321 submitted results. Frames per second are PassMark test-scene rates, not game frame rates.
Theoretical throughput
NVIDIA states a peak of AI performance 988 TOPS and Ray tracing 94 TFLOPS for the NVIDIA GeForce RTX 5070.
As stated by PassMark
PassMark's own listing for the NVIDIA GeForce RTX 5070 states the PCIe 5.0 x16 interface, up to 12 GB of memory, a core clock of 2330 MHz, DirectX 12_2 and OpenGL 4.6, first benchmarked on 2025-03-05. These are PassMark's figures, shown separately from the specification tables above.
Not to be confused with
This page covers the NVIDIA GeForce RTX 5070 specifically. It is a different product from NVIDIA GeForce RTX 5070 Laptop GPU, NVIDIA GeForce RTX 5070 Ti, NVIDIA GeForce RTX 5070 Ti Laptop GPU, which have their own specifications. Check the exact model name before comparing figures.
Graphics processor
The NVIDIA GeForce RTX 5070 is based on the Blackwell architecture, uses the GB205-300-A1 graphics processor, is manufactured on a 4 nm process, contains 31,100 million transistors and has a die size of 263 mm².
Shader configuration
The NVIDIA GeForce RTX 5070 has 6,144 shading units, 48 streaming multiprocessors, 192 texture mapping units, 80 render output units, 48 ray tracing cores and 192 tensor cores.
The NVIDIA GeForce RTX 5070 does not use XMX engines; those are specific to Intel Arc graphics. Infinity Cache is an AMD architectural feature and does not apply to the NVIDIA GeForce RTX 5070.
Clock speeds
The NVIDIA GeForce RTX 5070 runs at a base clock of 2.33 GHz and boosts to 2.51 GHz.
Memory
The NVIDIA GeForce RTX 5070 is equipped with 12 GB of GDDR7 memory, uses a 192-bit memory bus and delivers 672.0 GB/s of memory bandwidth.
Board and power
The NVIDIA GeForce RTX 5070 has a board power rating of 250 W, a total board power of 250 W, a power supply of at least 650 W is recommended, power is supplied through 2x PCIe 8-pin cables (adapter in box) OR 300 W or greater PCIe Gen 5 cable, the card occupies 2-slot, it is 242 mm long, its width is 112 mm, the maximum operating temperature is 85 °C and it connects over PCIe 5.0.
Market
The NVIDIA GeForce RTX 5070 was released on March 4, 2025 and launched at a price of $549.
Benchmarks
Graphics and compute APIs
The NVIDIA GeForce RTX 5070 supports DirectX 12 Ultimate, OpenGL 4.6, Vulkan 1.4 and CUDA compute capability 12.0.
Feex has no recorded OpenCL version for the NVIDIA GeForce RTX 5070.
Features
The NVIDIA GeForce RTX 5070 has 15 supported features recorded outside the media engine. A feature that no source mentions is shown as not listed, never as unsupported.
Rendering
-
Supported: DirectX 12 UltimateMicrosoft's top DirectX 12 feature tier, covering hardware ray tracing, mesh shaders, variable-rate shading and sampler feedback.
-
Supported: NVIDIA DLSS DLSS 4.5NVIDIA's AI-based upscaling and frame-generation technology that can raise frame rates in supported games.Modes: Super Resolution, DLAA, Ray Reconstruction, Frame Generation, Multi Frame Generation, Dynamic Multi Frame Generation
-
Supported: Hardware ray tracing 4th GenerationDedicated GPU hardware that speeds up ray-traced lighting, reflections and shadows in supported games and apps.Ray tracing units: 48
Not listed (1)
No source Feex holds states whether the NVIDIA GeForce RTX 5070 supports Vulkan ray tracing. That is silence, not a statement that they are unsupported.
The NVIDIA GeForce RTX 5070 supports DirectX 12 Ultimate, NVIDIA DLSS (DLSS 4.5) and Hardware ray tracing (4th Generation) under Rendering.
Compute and AI
-
Supported: CUDA Compute capability 12.0Support for NVIDIA's CUDA computing platform, used by many creative, scientific and AI applications.
-
Supported: Tensor cores 5th GenerationNVIDIA's dedicated matrix-math units used for AI workloads such as DLSS and machine-learning inference.Tensor cores: 192
Not listed (1)
No source Feex holds states whether the NVIDIA GeForce RTX 5070 supports NVLink / SLI. That is silence, not a statement that they are unsupported.
The NVIDIA GeForce RTX 5070 supports CUDA (Compute capability 12.0) and Tensor cores (5th Generation) under Compute and AI.
Display features
-
Supported: NVIDIA G-SYNCSupport for NVIDIA's variable refresh rate technology, which syncs the display to the GPU to reduce tearing and stutter.
-
Supported: HDCP content protection HDCP 2.3Support for HDCP, the copy-protection standard that some protected 4K streaming and disc video requires.
-
Supported: VR readyThe vendor states the GPU meets the requirements for running mainstream PC virtual reality headsets.
The NVIDIA GeForce RTX 5070 supports NVIDIA G-SYNC, HDCP content protection (HDCP 2.3) and VR ready under Display features.
Platform
-
Supported: Small-form-factor readyThe vendor lists the GPU as suitable for compact small-form-factor PC builds.
Not listed (1)
No source Feex holds states whether the NVIDIA GeForce RTX 5070 supports Resizable BAR. That is silence, not a statement that they are unsupported.
The NVIDIA GeForce RTX 5070 supports Small-form-factor ready under Platform.
Software and drivers
-
Supported: NVIDIA Game Ready driversDriver branch tuned and released alongside new games.
-
Supported: NVIDIA BroadcastNVIDIA software that applies AI noise removal and camera effects for calls and streams.
-
Supported: NVIDIA Reflex Reflex 2NVIDIA's latency-reduction technology that lowers the delay between input and what you see in supported games.
-
Supported: RTX HDRNVIDIA feature that adds HDR to games and video that were made without it.
-
Supported: RTX RemixNVIDIA tooling for modding older games with ray tracing and updated assets.
-
Supported: NVIDIA Studio driversDriver branch tuned and tested for creative applications.
Not listed (3)
No source Feex holds states whether the NVIDIA GeForce RTX 5070 supports NVIDIA Freestyle, NVIDIA Omniverse and NVIDIA ShadowPlay. That is silence, not a statement that they are unsupported.
The NVIDIA GeForce RTX 5070 supports NVIDIA Game Ready drivers, NVIDIA Broadcast, NVIDIA Reflex (Reflex 2), RTX HDR, RTX Remix and NVIDIA Studio drivers under Software and drivers.
Media engine
The NVIDIA GeForce RTX 5070 provides hardware encode for H.264 (AVC), HEVC (H.265) and AV1 and hardware decode for H.264 (AVC), HEVC (H.265), AV1 and VP9.
| Codec | Encode | Decode |
|---|---|---|
| H.264 (AVC) | Supported4:4:4: Yes | Supported |
| HEVC (H.265) | Supported4:2:2: Yes; 4:4:4: Yes; 10-bit: Yes; B-frames: Yes; 8K: Yes | Supported4:4:4: Yes |
| AV1 | Supported | Supported |
| VP9 | Not stated by the vendor | Supported |
supported explicitly not supported not stated by the vendor
Media engines recorded for the NVIDIA GeForce RTX 5070: NVIDIA NVENC encoder (9th Generation, encoder engines: 1) and NVIDIA NVDEC decoder (6th Generation, decoder engines: 1).
Display outputs
The NVIDIA GeForce RTX 5070 offers outputs 3x DisplayPort, 1x HDMI, up to 4 simultaneous displays, a maximum resolution of 4K at 480Hz or 8K at 165Hz with DSC and HDCP 2.3. It is stated to be VR ready.
Frequently asked questions
How much VRAM does the NVIDIA GeForce RTX 5070 have?
The NVIDIA GeForce RTX 5070 has 12 GB of GDDR7 memory on a 192-bit bus. Its memory bandwidth is 672.0 GB/s.
How much power does the NVIDIA GeForce RTX 5070 use and what power supply does it need?
The NVIDIA GeForce RTX 5070 has a board power rating of 250 W. Feex records a suggested power supply of at least 650 W. Power connectors: 2x PCIe 8-pin cables (adapter in box) OR 300 W or greater PCIe Gen 5 cable.
How fast is the NVIDIA GeForce RTX 5070?
The NVIDIA GeForce RTX 5070 scores 28,677 in the PassMark G3D benchmark, higher than 99.1% of the 2,658 GPUs benchmarked. Its PassMark rank is #27. In the PassMark DirectX 12 test it averages 106 fps.
Does the NVIDIA GeForce RTX 5070 support ray tracing?
Yes. The NVIDIA GeForce RTX 5070 has hardware ray tracing (4th Generation) with 48 ray tracing cores.
Does the NVIDIA GeForce RTX 5070 support AV1 encoding?
Yes. The NVIDIA GeForce RTX 5070 supports AV1 hardware encoding and AV1 hardware decoding.
Does the NVIDIA GeForce RTX 5070 support DirectX 12 Ultimate?
Yes. The NVIDIA GeForce RTX 5070 supports DirectX 12 Ultimate.
What is the CUDA compute capability of the NVIDIA GeForce RTX 5070?
The NVIDIA GeForce RTX 5070 has CUDA compute capability 12.0.
What bus interface does the NVIDIA GeForce RTX 5070 use?
The NVIDIA GeForce RTX 5070 connects over PCIe 5.0.
When was the NVIDIA GeForce RTX 5070 released and how much did it cost at launch?
The NVIDIA GeForce RTX 5070 was released on March 4, 2025 and launched at a price of $549.
Corrections to this record
After an audit, 2 values on the NVIDIA GeForce RTX 5070 record were corrected. Each change is listed with its old value, new value, rule and sources.
- Base clock:
2,160 MHz→ 2,325 MHz. Rule: Two independent sources agree. 2 independent sources agree on 2325 and contradict the stored 2160. Sources: Wikipedia: List of Nvidia graphics processing units (revision 1377107724); Wikipedia: List of Nvidia desktop graphics processing units (revision 1375708460). - Compute units:
48cleared (now Not recorded). Rule: NVIDIA SM count moved out of compute units. compute units 48 on an NVIDIA GPU is its SM count (stated by the Wikipedia list articles). Sources: Wikipedia: List of Nvidia graphics processing units (revision 1377107724); Wikipedia: List of Nvidia desktop graphics processing units (revision 1375708460).
The audit log is public: all changes for this GPU, the corrections overview and the method.
Similar performance
Compare the NVIDIA GeForce RTX 5070
Side-by-side specification comparisons against cards in the same performance band.