Rent a GPU

Can the AMD Radeon RX 9070 run gpt-oss 120B?

Radeon RX 9070 × gpt-oss 120B8K context
NoDoes not fit in memory
Memory at Q4_K_M71.7 GB of 16 GB
Weights 70.8 GBKV cache 0.3 GBRuntime 0.6 GB
On the GPU21%of the Q4_K_M weights
In system RAM55.7 GBthe rest spills over
Offload speed~14tokens per second
  1. FP16
  2. Q8_0
  3. Q6_K
  4. Q5_K_M
  5. Q4_K_M
  6. Q3_K_M
  7. Q2_K

No. Even at Q4_K_M, gpt-oss 120B needs about 71.7 GB, but the AMD Radeon RX 9070 has 16 GB usable. You could offload some layers to system RAM, but it will be very slow.

Try other context lengths

Run it on a rented GPU

On a cloud GPU with enough memory, the same command you'd use locally works:

ollama run gpt-oss:120b

At Q4_K_M it needs about 71.7 GB, so pick a GPU with at least 72 GB.

New to this? Step-by-step setup for the Radeon RX 9070

Run it in the cloud

Too big for the AMD Radeon RX 9070? Run gpt-oss 120B on a rented GPU instead.

Rent a cloud GPU by the hour. No upfront cost, stop any time. Not sure whether to rent or buy? Compare the cost.

Rent a GPU on RunPodRent a GPU on Vast.ai

Memory needed by quantization

At 8,192 tokens of context. The black line marks the 16 GB the AMD Radeon RX 9070 can use.

QuantQualityMemoryFits?Speed
FP16Full quality234.5 GBToo big—
Q8_0Near-lossless125 GBToo big—
Q6_KExcellent96.7 GBToo big—
Q5_K_MVery good84 GBToo big—
Q4_K_MGood, most popular71.7 GBToo big—
Q3_K_MNoticeably degraded58 GBToo big—
Q2_KHeavily degraded44.7 GBToo big—

Measured speed from real Radeon RX 9070 owners

Tokens per second that people measured on their own AMD Radeon RX 9070, next to our estimate.

No one has measured gpt-oss 120B on the AMD Radeon RX 9070 yet. The speeds above are estimates. Run the benchmark and yours becomes the first real number on this page.

Have a Radeon RX 9070? Add your real speed

One command times gpt-oss 120B in your own Ollama, shows you exactly what it would send, and asks before sending anything. No sign-up, no personal data.

macOS and Linux

curl -fsSL https://gpufit.dev/bench.sh | sh -s -- gpt-oss:120b

Windows (PowerShell)

$env:GPUFIT_MODEL="gpt-oss:120b"; irm https://gpufit.dev/bench.ps1 | iex

How the benchmark works

GPUs that run gpt-oss 120B comfortably

Check RTX PRO 6000 Blackwell prices

Models that run well on the AMD Radeon RX 9070

Frequently asked questions

Can AMD Radeon RX 9070 run gpt-oss 120B?

No. Even at Q4_K_M, gpt-oss 120B needs about 71.7 GB, but the AMD Radeon RX 9070 has 16 GB usable. You could offload some layers to system RAM, but it will be very slow.

How much VRAM does gpt-oss 120B need?

About 71.7 GB at Q4_K_M with 8,192 tokens of context, and about 234.5 GB at full FP16 precision.

How fast is gpt-oss 120B on the AMD Radeon RX 9070?

It does not fit in the AMD Radeon RX 9070's memory, so most layers would run from system RAM and generation would be very slow.