PNY XLR8 RTX 4070 VERTO EPIC-X RGB Triple Fan
The PNY XLR8 RTX 4070 VERTO EPIC-X RGB Triple Fan is PNY’s version of the GeForce RTX 4070. At 305 mm it’s 65 mm longer than the 240 mm reference design. It uses 1 8-pin rather than the reference 1 16-pin. Below you’ll find every current retailer price and the full spec sheet.
Prices update regularly and may differ at the retailer. This page contains affiliate links.
Gaming performance
Gaming score: geometric mean of in-game FPS, normalized so the RTX 5090 = 100. Render and synthetic scores are in the Benchmarks section below.
Gaming scores are normalised performance indexes derived from Tom’s Hardware’s published GPU benchmarks. Performance per dollar and per watt are our own calculations on top of that data.
1080p Gaming
1440p Gaming
4K Gaming
The GeForce RTX 4070 scores 65.6 at 1080p, 47.9 at 1440p, and 37.2 at 4K on a scale where 100 = the fastest GPU we test (RTX 5090). These scores apply to all board versions of this chip.
Benchmarks
Standardized benchmark scores for this chip, each compared to the fastest card we track and linked to its source. They apply to every board version.
3D & gaming
Render
Compute & AI
Where to buy
No in-stock prices right now — check back soon.
No in-stock prices right now
We’re not tracking a live in-stock price for this exact version at the moment. Prices refresh daily — check back soon, or compare other versions of the GeForce RTX 4070 below.
Specifications
GPU chip
- Chip
- GeForce RTX 4070
- Architecture
- Ada Lovelace AD104
- Process node
- TSMC 4N
- Shader cores
- 5,888
- SM count
- 46
- Texture units (TMUs)
- 184
- Render outputs (ROPs)
- 64
- RT cores
- 46 · 3rd gen
- Tensor / AI cores
- 184 · 4th gen
Memory
- VRAM
- 12GB GDDR6X
- Memory bus
- 192-bit
- Bandwidth
- 504 GB/s
- Memory speed
- 21 Gbps
- L2 cache
- 36 MB
GPU die
- Foundry
- TSMC
- Die size
- 295 mm²
- Transistors
- 35.8 billion
- Generation
- GeForce RTX 40
Display & connectivity
- Max resolution
- 7680 × 4320
- Max displays
- 4
- HDCP
- 2.3
APIs & software
- DirectX
- 12 Ultimate (12.2)
- Vulkan
- 1.3
- OpenGL
- 4.6
- OpenCL
- 3.0
- CUDA
- Compute 8.9
- Shader model
- 6.8
Platform features
- Upscaling
- DLSS Super Resolution, Ray Reconstruction, Frame Generation
- Also runs
- FSR 3, FSR 3 Frame Generation, XeSS 2, XeSS 2 Frame Generation
- Resizable BAR
- Yes
Video encoding & decoding
| Codec | Encode | Decode |
|---|---|---|
| H.264 / AVC | Supported | Supported |
| H.265 / HEVC | Supported | Supported |
| AV1 | Supported | Supported |
| VP9 | Not supported | Supported |
Encoder: NVENC 8th gen · Decoder: NVDEC 5th gen
Board
- Boost clock
- 2475 MHz(reference)
- Base clock
- 1920 MHz
- TDP
- 200W
- Recommended PSU
- 650W
- Power connectors
- Reference design:
1 16-pinThis card: 1 8-pin
Physical
- Cooling
- Air
- Warranty
- 3 years
- Length
- Reference design:
240 mmThis card: 305 mm - Slot width
- 2 slots
- Outputs
- 1 HDMI 2.1, 3 Displayport 1.4a
Chip-level specs — cores, memory, die, APIs, media engine and platform features — are shared by every version of the GeForce RTX 4070. The Board, Physical and Identifier boxes are specific to this PNY variant; a struck-through value shows where it differs from the reference design.
Local AI performance
What the PNY XLR8 RTX 4070 VERTO EPIC-X RGB Triple Fan can run on your own machine, and how fast. Capacity and speed come from the GeForce RTX 4070 chip, so they are the same on every board version; price per GB is this board’s own.
Largest model it can run (8K context)
- Q4 (the usual choice, near-original quality)
- ~14B
- Q8 (closer to original, half the model size)
- ~8.9B
- FP16 (unquantised, for training and research)
- ~4.7B
Parameter counts the card holds with room for an 8K conversation, counting the model weights plus its attention cache. A longer conversation needs more memory, so it lowers these figures.
Speed & value
- Speed, 8B model at Q4
- 76-89 tok/s
- Memory bandwidth
- 504 GB/s
- Price per GB of VRAM
- —
Generation speed tracks memory bandwidth, because every word produced reads the whole model from memory. Shown as a range.
Models that fit, at Q4
| Model | Size | Memory needed | Est. speed |
|---|---|---|---|
| Starcoder2 15B | 16.0B | 9.7GB | 41-48 tok/s |
| Sqlcoder 15B | 15.0B | 10.1GB | 41-48 tok/s |
| Cogito 14B | 14.8B | 10.1GB | 42-48 tok/s |
Popular models small enough for this card at Q4, one per family. Speed is left blank for mixture-of-experts models, where a bandwidth estimate is not reliable enough to publish.
Choosing a card for AI? Compare every GPU for local AI, including setups that pool the memory of two or more cards.
Across generations
Where the GeForce RTX 4070 sits in PNY’s lineup — the generation before and after.
Other versions of the GeForce RTX 4070
Same GPU, different board, picked by what buyers usually compare.
Affiliate disclosure: Some links on this page are affiliate links — MaxMyBuild may earn a commission at no extra cost to you if you buy through them. As an Amazon Associate, MaxMyBuild earns from qualifying purchases. For Amazon, current price and availability are shown on the Amazon product page; product prices and availability are accurate as of the date and time of purchase and are subject to change. Prices shown for other retailers are approximate and may differ at checkout. “Search” links are non-affiliate and provided for convenience only.