Can the Radeon RX 9070 XT run Llama 4 Scout (17B x 16E) Instruct?
Does not fit at Q4_K_M.
Llama 4 Scout (17B x 16E) Instruct is 109B parameters (MoE; all experts resident). At Q4_K_M the weights are about 60.9 GiB; with the 1.5 GiB runtime reserve that is 62.4 GiB against the card's 16GB, about 46.4 GiB more than the card has.
All three quants
16GB card| Quant | Weights | With reserve | Verdict | Headroom |
|---|---|---|---|---|
| Q4_K_Mthe everyday quant | 60.9 GiB | 62.4 GiB | Does not fit | short 46.4 GiB |
| Q8_0near-lossless | 107.6 GiB | 109.1 GiB | Does not fit | short 93.1 GiB |
| FP16full weights | 203.0 GiB | 204.5 GiB | Does not fit | short 188.5 GiB |
Verdict rules: fits comfortably when weights plus reserve sit within 85% of VRAM; tight when they fit with little left for context; CPU offload when they exceed VRAM by up to 2x (runs, slowly, with layers in system RAM); no beyond that. The KV cache per token depends on the model's architecture and is not modelled here.
The numbers for this pairing
At Q4_K_M the weights take about 60.9 GiB, so with the 1.5 GiB runtime reserve it needs 62.4 GB: 46.4 GB more than the card's 16 GB.
No version of it fits entirely in this card's memory, so it would lean on system RAM and run far slower.
It is a mixture-of-experts model: about 17B of its 109B parameters work on each token, which helps speed, but all of them still have to fit in memory.
At about $810, this card costs roughly $51 per GB of VRAM.
Running it at its full 304 W board power for 4 hours a day would use about 36 kWh a month, about $6.71 at the US average residential rate of 18.4 cents per kWh (2026-05); real use sits below that ceiling.
Prices are the lowest current listings our tracker saw on Oct 7, 2026, since averages include premium variants; launch prices are marked where we have no listing.
Not this card. These fit it comfortably at Q4.
Nothing in the consumer catalog fits Llama 4 Scout (17B x 16E) Instruct at Q4 on a single card; this is multi-GPU or datacenter territory, or a job for hosted inference (see PicksByModel for per-token pricing).
What the Radeon RX 9070 XT does run well → · Llama 4 Scout (17B x 16E) Instruct on every card →
About the model
Llama 4 Scout (17B x 16E) Instruct by Meta: 10,485,760-token native context, text + vision, released 2025 under the Llama 4 Community License. Model card → · Hosted pricing on PicksByModel →