GeForce RTX 4060 Price, 8GB VRAM & Local LLMs
Price alert
We'll email you when the eBay (US) price reaches your target.
NVIDIA's RTX 4060 (2023) is an entry-level gaming card with 8GB of VRAM, suited to models up to 9B at 4-bit, while a 12B has to drop to 2-bit. The card draws only 115W, and NVIDIA recommends a 550W power supply for the system.
eBay
GeForce RTX 4060 price history
eBay
As an eBay Partner, VRAMGlass may be compensated if you buy through the eBay links on this page. The price you pay stays the same.
GeForce RTX 4060 VRAM & full specs
- Memory8 GB · GDDR6
- Bandwidth272 GB/s
- FP16 / BF1660.5 TFLOPS
- FP8 / INT8121 / 121 TOPS
- Bus width
- 128-bit
- Architecture
- Ada Lovelace (AD107)
- Cores
- 3,072
- Tensor cores
- 96
- Ecosystem
- CUDA
- Interface
- PCIe 4.0 x8
- NVLink
- Not supported
- Power
- 115 W
- Power connector
- 1× 8-pin
- Type
- Consumer
- Launch date
- Jun 29, 2023
- MSRP
- $299
Best GeForce RTX 4060 deals
Powered by VRAMGlass real-time price tracking — handpicking the best-value hardware deals across the web
As an eBay Partner, VRAMGlass may be compensated if you buy through the eBay links on this page. The price you pay stays the same.
UsedZOTAC Gaming GeForce RTX 4060 8GB Twin Edge 128 Bit GDDR6 Graphics GPU TESTED$349+ shippingBuy on eBay
UsedUSED! MSI RTX 4060 8GB VENTUS 2X BLACK OC VIDEO CARD$350Buy on eBay
UsedZotac GeForce RTX 4060 Twin Edge OC 8GB GDDR6 Graphics Card ZT-D40600E-10MTop Rated$370Buy on eBay
UsedGigabyte GeForce RTX 4060 Windforce OC 8GB GDDR6 Graphics Card GV-N4060WF2OC-8GDTop Rated$370Buy on eBay
UsedZotac GeForce RTX 4060 Twin X2 OC 8GB GDDR6 Graphics Card N40602-08D6X-173051WTop Rated$380Buy on eBay
UsedMSI GeForce RTX 4060 Ventus 2X OC 8GB GDDR6 Graphics CardTop Rated$385Buy on eBay
As an Amazon Associate I earn from qualifying purchases. The price you pay stays the same.
UsedDual GeForce RTX™ 4060 V2 OC Edition (PCIe 4.0, 8GB GDDR6, DLSS 3, HDMI 2.1a, DisplayPort 1.4a, 2-Slot Design, Axial-tech Fan Design, 0dB Technology, and More) (Renewed)$470Buy on Amazon
UsedGaming GeForce RTX 4060 8GB GDRR6 128-Bit HDMI/DP Nvlink TORX Fan 4.0 Ada Lovelace Architecture Graphics Card (RTX 4060 Ventus 2X Black 8G OC) (Renewed)$480$720-33%Buy on Amazon
UsedGeForce RTX 4060 WINDFORCE OC 8G Graphics Card, 2X WINDFORCE Fans, 8GB 128-bit GDDR6, GV-N4060WF2OC-8GD Video Card (Renewed)$500$600-17%Buy on Amazon
What LLMs can a single GeForce RTX 4060 run?
A single RTX 4060 is best with models up to 9B at 4-bit, running Qwen3.5 9B at about 34 tokens/s with 54K of context. With experts in system RAM, a 30B-class MoE at 2-bit actually runs faster than that 9B does on the card, as long as your RAM can hold it.
| Model | Size | Runs? | Speed t/s | Max context |
|---|---|---|---|---|
 Gemma 4 E4B8B | 3.8 GB | Runs | 50 | 128K |
 Qwen3.5 9B9.7B | 4.1 GB | Runs | 46.1 | 101K |
| Gemma 4 12B12B | 4.7 GB | Runs | 39.9 | 147K |
| Gemma 4 26B A4B25.2B (3.8B active) | 10.5 GB | Needs RAM offload | 45.7 | 250K |
 GLM 4.7 Flash30B (3B active) | 11.9 GB | Needs RAM offload | 45.7 | 107K |
| Qwen3.6 35B A3B36B (3B active) | 12.3 GB | Needs RAM offload | 65.2 | 256K |
| Model | Size | Runs? | Speed t/s | Max context |
|---|---|---|---|---|
| Gemma 4 E4B8B | 5 GB | Runs | 38.8 | 128K |
| Qwen3.5 9B9.7B | 5.7 GB | Runs | 34 | 54K |
| Gemma 4 12B12B | 7.1 GB | Too big | — | — |
| Gemma 4 26B A4B25.2B (3.8B active) | 16.9 GB | Needs RAM offload | 31.4 | 221K |
| GLM 4.7 Flash30B (3B active) | 18.3 GB | Needs RAM offload | 32.7 | 102K |
| Qwen3.6 35B A3B36B (3B active) | 22.1 GB | Needs RAM offload | 42.5 | 250K |
| Model | Size | Runs? | Speed t/s | Max context |
|---|---|---|---|---|
| Gemma 4 E4B8B | 8.2 GB | Too big | — | — |
| Qwen3.5 9B9.7B | 9.5 GB | Too big | — | — |
| Gemma 4 12B12B | 12.7 GB | Too big | — | — |
| Gemma 4 26B A4B25.2B (3.8B active) | 26.9 GB | Needs RAM offload | 21.1 | 176K |
| GLM 4.7 Flash30B (3B active) | 31.8 GB | Needs RAM offload | 20.4 | 93K |
| Qwen3.6 35B A3B36B (3B active) | 36.9 GB | Needs RAM offload | 27.9 | 212K |
RunsFits fully in VRAM; speed is estimated from VRAM bandwidth.
Needs RAM offloadMoE only: expert weights sit in system RAM, so speed is estimated from 70 GB/s RAM bandwidth.
Too bigWon’t fit on one card at this quantization, even with RAM offload.
Full calculation rules are on the Methodology page.
Who it suits
Good fit
- Everyday use with small models
Gemma 4 E4B at 4-bit takes just 5 GB, so you can open the full 128K of context
- You want to try 30B-class MoE models
Qwen3.6 35B A3B at 2-bit, with its experts in system RAM, runs at about 65 tokens/s, and its speed depends mostly on your RAM
- Adding a GPU to an office PC or small desktop
It draws little power, and one 8-pin connector is enough
Not a good fit
- Generation speed matters to you
With only 272 GB/s of memory bandwidth, it generates noticeably slower on the same model
- You want to run 12B or larger
Gemma 4 12B at 4-bit needs 7.1 GB, and with context added it won't fit in 8GB; 27B-class dense models don't fit at all
Before you buy
Less bandwidth than the 3060 12GB
When running models, generation speed depends almost entirely on memory bandwidth. The 4060's memory bus is cut to 128-bit, giving 272 GB/s, lower than the 360 GB/s of the previous-generation RTX 3060 12GB, and it has 4GB less VRAM too. Its strengths are low power draw and little heat.
Don't mix up the 4060 and 4060 Ti
The 4060 Ti, with the extra “Ti” in its name, comes in 8GB and 16GB versions, with more cores and 288 GB/s of bandwidth; if you want twice the VRAM, look at the RTX 4060 Ti 16GB. If a listing just says “4060”, ask the seller for a GPU-Z screenshot and check that the core count is 3072.
The laptop 4060 is a different story
The RTX 4060 Laptop GPU has the same 3072 cores and 8GB as the desktop card, but its power limit is set by the laptop maker, so you can't estimate how fast it runs models from this page.
Compared with similar GPUs
GeForce RTX 4060 vs GeForce RTX 5060, GeForce RTX 3060 12GB, and GeForce RTX 4060 Ti 16GB: price, VRAM, bandwidth, compute and local LLM speed compared.
| Spec | GeForce RTX 4060 | GeForce RTX 5060 | GeForce RTX 3060 12GB | GeForce RTX 4060 Ti 16GB |
|---|---|---|---|---|
| Memory | 8 GB | 8 GBSame | 12 GB50% more | 16 GB100% more |
| Bandwidth | 272 GB/s | 448 GB/s65% higher | 360 GB/s32% higher | 288 GB/s6% higher |
| BF16 / FP16 compute | 61 TFLOPS | 77 TFLOPS27% higher | 51 TFLOPS16% lower | 88 TFLOPS46% higher |
| INT8 compute | 121 TOPS | 154 TOPS27% higher | 102 TOPS16% lower | 177 TOPS46% higher |
| Qwen3.8 27B speed (2-bit) | Not supported | Not supported | 26.4 t/s | 21.3 t/s |
| Largest dense model (4-bit) | 10B | 10BSame | 16B60% more | 23B130% more |
| Power | 115 W | 145 W26% higher | 170 W48% higher | 165 W43% higher |
| Interface | PCIe 4.0 x8 | PCIe 5.0 x8 | PCIe 4.0 x16 | PCIe 4.0 x8 |
| Released | 2023-06 | 2025-05 | 2021-02 | 2023-07 |
| eBay median |
As an eBay Partner, VRAMGlass may be compensated if you buy through the eBay links on this page. The price you pay stays the same.
Amazon.jp
Xianyu