Can the GeForce RTX 3090 (used) run gpt-oss-120b?
Does not fit at Q4_K_M.
gpt-oss-120b is 117B parameters (MoE; all experts resident). At Q4_K_M the weights are about 65.4 GiB; with the 1.5 GiB runtime reserve that is 66.9 GiB against the card's 24GB, about 42.9 GiB more than the card has.
All three quants
24GB card| Quant | Weights | With reserve | Verdict | Headroom |
|---|---|---|---|---|
| Q4_K_Mthe everyday quant | 65.4 GiB | 66.9 GiB | Does not fit | short 42.9 GiB |
| Q8_0near-lossless | 115.5 GiB | 117.0 GiB | Does not fit | short 93.0 GiB |
| FP16full weights | 217.9 GiB | 219.4 GiB | Does not fit | short 195.4 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 65.4 GiB, so with the 1.5 GiB runtime reserve it needs 66.9 GB: 42.9 GB more than the card's 24 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 5.1B of its 117B parameters work on each token, which helps speed, but all of them still have to fit in memory.
At about $1,800, this card costs roughly $75 per GB of VRAM.
Running it at its full 350 W board power for 4 hours a day would use about 42 kWh a month, about $7.73 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 gpt-oss-120b 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 GeForce RTX 3090 (used) does run well → · gpt-oss-120b on every card →
About the model
gpt-oss-120b by OpenAI: 131,072-token native context, text, released 2025 under the Apache-2.0. Model card → · Hosted pricing on PicksByModel →