Turing GPU without Tensor Cores.
Weekly median price per GPU per hour · Get the data
By provider shows one card per company, placed where its best offer ranked. By configuration lists every offer.
What you can rent comes first: in stock, then waitlist, then not reported, then out of stock. Priced offers rank ahead of quote-only, and on-demand ahead of other billing types.
Within a group, five factors set the order:
We match the provider, its country, GPU model, form factor, billing type, availability and instance name. Matching is partial and case-insensitive. Everyday words work too, so "interruptible" finds spot and "sold out" finds out of stock.
No providers match .
Heads up: A provider's own page may quote a different figure, for example on tax, region or a promotion. A monthly-only plan shows a derived hourly rate. Verify before provisioning. More on how we price.
One GTX 1660 has 6 GB of VRAM. Even at 4-bit, the weights of most modern AI models exceed the card's usable VRAM, and context adds KV cache on top. Below are some open-weight LLMs, with the estimated memory and GPUs each one needs.
| Model | Memory (INT4 / FP4) | GTX 1660s needed | Cost /hr | Cost /mo |
|---|---|---|---|---|
Qwen3.8-27B
27B
|
19 GB
|
4
|
–
|
–
|
Gemma 4 31B
30.7B
|
24 GB
|
5
|
–
|
–
|
GPT-OSS-120B
117B · 5.1B active
|
68 GB
|
13
|
–
|
–
|
GLM-5.3-Flash
320B · 18B active
|
178 GB
|
34
|
–
|
–
|
MiniMax-M3
428B · 23B active
|
241 GB
|
45
|
–
|
–
|
DeepSeek V4.1 Flash
552B · 16B active
|
306 GB
|
57
|
–
|
–
|
Kimi K3
2.8T · 104B active
|
1,545 GB
|
287
|
–
|
–
|
Estimates based on the median on-demand rate. Memory is weights plus FP16/BF16 KV cache at 32K context per request. GPU counts assume 90% of advertised VRAM is usable. No guarantee of runtime support, usable performance, or that a matching quantized build exists. Pricing methodology.
Nvidia GTX 1660 · Per GPU
| Compute · dense | |
|---|---|
| FP16 / BF16 | 10.1 TFLOPS |
| FP32 | 5.0 TFLOPS |
| Precision support | FP16FP32 |
| Memory | |
|---|---|
| Capacity | 6 GB GDDR5 |
| Bandwidth | 192 GB/s |
| Bus width | 192-bit |
| ECC | No |
| Silicon | |
|---|---|
| Architecture | Turing |
| Process | TSMC 12 nm FinFET |
| Transistors | 6.6 billion |
| Shader cores | 1,408 CUDA cores |
| Compute units | 22 SMs |
| Fabric and host | |
|---|---|
| Host interface | PCIe 3.0 x16 |
| Power | |
|---|---|
| Board power | 120 W |
| Cooling | Active |
Source: official Nvidia GTX 1660 datasheet.
As of October 3, 2026, we track 8 configs from 3 providers. Prices are per GPU per hour.
| Billing type | Configs | Cheapest |
|---|---|---|
On-demand | 2 | $0.04 (in stock, Salad) |
Reserved | 4 | $0.19 (24 mo) |
Spot | 2 | $0.02 (in stock, Salad) |
No median for on-demand, reserved and spot: we only show this when at least 3 providers list the GPU on that billing type.
The cheapest verified in-stock estimate is $29 per month on-demand, $14 spot.
GetDeploying currently tracks GTX 1660 configs from 3 providers. Salad and Vast.ai have verified in-stock on-demand configs. See the full price comparison above for every provider and config.
More than one: even at 4-bit, the weights of most modern AI models exceed one GTX 1660's usable VRAM, and context adds KV cache on top. Larger models run across multiple GPUs: the model table above shows the estimated memory and GPU count for popular open-weight LLMs.
6GB GDDR5. Budget Turing GPU without RT or Tensor Cores.
No Tensor Cores and only 6GB VRAM. Only viable for basic cloud desktop rendering or legacy compute tasks.
Nvidia RTX 3060
12 GB GDDR6 · 88% more bandwidth than the GTX 1660
Nvidia RTX 2060
6 GB GDDR6 · 75% more bandwidth than the GTX 1660
Nvidia GTX 1660 Super
6 GB GDDR6 · 75% more bandwidth than the GTX 1660
Last updated
