High-end RDNA 4 consumer GPU for local AI inference and gaming.
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 Radeon RX 9070 XT has 16 GB of VRAM. In practice, that's enough memory for roughly 15B parameters at 4-bit or 2B at 16-bit, assuming a 32K context. Below are some open-weight LLMs, with the estimated memory and GPUs each one needs.
| Model | Memory (INT4 / FP4) | Radeon RX 9070 XTs needed | Cost /hr | Cost /mo |
|---|---|---|---|---|
Qwen3.8-27B
27B
|
18 GB
|
2
|
–
|
–
|
Gemma 4 31B
30.7B
|
22 GB
|
2
|
–
|
–
|
GPT-OSS-120B
117B · 5.1B active
|
67 GB
|
5
|
–
|
–
|
GLM-5.3-Flash
320B · 18B active
|
178 GB
|
13
|
–
|
–
|
MiniMax-M3
428B · 23B active
|
239 GB
|
17
|
–
|
–
|
DeepSeek V4.1 Flash
552B · 16B active
|
306 GB
|
22
|
–
|
–
|
Kimi K3
2.8T · 104B active
|
1,544 GB
|
108
|
–
|
–
|
Estimates based on the median on-demand rate. Memory is weights plus FP8 KV cache at 32K context per request (FP16/BF16 in the 16-bit column). 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.
AMD Radeon RX 9070 XT · Per GPU
| Compute · dense | |
|---|---|
| FP8 | 389 TFLOPS 779 with sparsity |
| FP16 / BF16 | 195 TFLOPS 389 with sparsity |
| INT8 | 389 TOPS 779 with sparsity |
| FP32 | 48.7 TFLOPS |
| Precision support | FP8FP16BF16FP32INT4INT8 |
| Memory | |
|---|---|
| Capacity | 16 GB GDDR6 |
| Bandwidth | 640 GB/s |
| Bus width | 256-bit |
| Silicon | |
|---|---|
| Architecture | RDNA 4 |
| Process | 4nm |
| Transistors | 53.9 billion |
| Shader cores | 4,096 Stream processors |
| Matrix cores | 128 |
| Compute units | 64 |
| Fabric and host | |
|---|---|
| Host interface | PCIe 5.0 x16 |
| Power | |
|---|---|
| Board power | 304 W |
As of November 13, 2026, we track 4 configs from 1 provider. Prices are per GPU per hour.
| Billing type | Configs | Cheapest |
|---|---|---|
On-demand | 1 | $0.20 (in stock, Salad) |
Spot | 3 | $0.07 (not verified in stock, Salad) |
No median for on-demand 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 $144 per month on-demand, $79 spot.
GetDeploying currently tracks Radeon RX 9070 XT configs from 1 provider. Salad has verified in-stock on-demand configs. See the full price comparison above for every provider and config.
One Radeon RX 9070 XT runs models up to roughly 15B parameters at 4-bit quantization or 2B at 16-bit, assuming a 32K context. Larger models run across multiple GPUs: the model table above shows the estimated memory and GPU count for popular open-weight LLMs.
16GB on RDNA 4, with FP8 matrix support and structured sparsity that the RDNA 3 cards lack. Holds a 14B model at 8-bit or a 7B model at FP16 on one card, and AMD lists it as supported by ROCm on Linux.
Models above about 14B parameters at 8-bit do not fit in 16GB without heavier quantization. Where a framework needs CUDA, the RTX 5070 Ti is the 16GB consumer counterpart. The ROCm software ecosystem, while rapidly improving, is narrower than CUDA.
Nvidia RTX 5070 Ti
16 GB GDDR7 · 40% more bandwidth than the Radeon RX 9070 XT
AMD Radeon RX 7900 XTX
24 GB GDDR6 · 50% more bandwidth than the Radeon RX 9070 XT
AMD Radeon RX 9060 XT
16 GB GDDR6 · 50% less bandwidth than the Radeon RX 9070 XT
Last updated
