Home / Value / Cheapest GPU for GPT-OSS 20B
Cheapest GPU for GPT-OSS 20B
Q4 · 8K (default)
Ranked by lowest fresh Amazon-matched board-partner SKU price among GPUs that fully fit this exact workload. This is not a speed ranking and not a “best GPU” list. Query combinations are not indexed.
To open another model, use the value explorer or the model catalog.
Requirement
- Model
- GPT-OSS 20B
- Quantization · context
- Q4 · 8K
- Required VRAM
- 14.7 GB
- Calculator
- v1.0.0
23 canonical GPUs fully fit. 14 of them have a fresh current price. 9 compatible GPUs have no current price.
Formula: MIN(live fresh Amazon-matched board-partner SKU price) WHERE compatibility = FITS_IN_VRAM. Sort: price ASC, advertised VRAM ASC, canonical GPU id ASC. Prices are the lowest fresh Amazon-matched board-partner SKU, not MSRP or a street-price average.
Among 14 canonical GPUs with a fresh matched Amazon offer currently available in RigForAI. Ranked 14. Unpriced compatible: 9. This is not the entire US market — only canonical GPUs with a fresh matched Amazon offer currently in RigForAI.
Cheapest currently available full-VRAM GPU
NVIDIA Quadro P6000 24 GB
- Advertised VRAM
- 24 GB
- Usable VRAM
- 21.6 GB
- Headroom
- 6.9 GB
- Compatibility
- FITS_IN_VRAM
- Current fresh Amazon price
- $646.96
- Winning physical SKU
- HP NVIDIA Quadro P6000 (24GB) Graphics CardHPASIN B01NALQ0PB
- Offer fetched
- 2026-08-14 20:19 UTC
Why it ranks first
- FITS_IN_VRAM for this model, quantization, and context
- current fresh matched Amazon offer (EXACT/HIGH, not a bundle)
- lowest price among 14 eligible priced GPUs
- price ASC, advertised VRAM ASC, canonical GPU id ASC.
Next option: AMD Radeon RX 7900 XTX 24 GB $1059.99 · difference +$413.03. Build a complete PC around this option. GPU-only price is not a complete-build cost.
Other currently priced full-fit GPUs
Ranked by lowest fresh Amazon-matched board-partner SKU price among GPUs that fully fit this exact workload.
Compatible — current price unavailable
These GPUs calculate as a full VRAM fit. They are not inserted into the current-price ranking because RigForAI has no fresh matched Amazon offer for them. That is not the same as too expensive or incompatible.
- NVIDIA RTX 4500 24 GB · 24 GB advertised
- NVIDIA Tesla K80 24 GB · 24 GB advertised
- NVIDIA Tesla M10 32 GB · 32 GB advertised
- NVIDIA Tesla M40 24 GB · 24 GB advertised
- NVIDIA Quadro GV100 32 GB · 32 GB advertised
- NVIDIA RTX 6000 24 GB · 24 GB advertised
- NVIDIA RTX 8000 48 GB · 48 GB advertised
- NVIDIA RTX A6000 48 GB · 48 GB advertised
- NVIDIA RTX A5500 24 GB · 24 GB advertised
Supported llama.cpp decode per current $1,000
Separate ranking. Same model, quantization, llama.cpp CUDA, llama_bench_pp512_tg128, published decode prediction, confidence HIGH or MEDIUM, FITS_IN_VRAM, fresh price. Formula: decode tok/s / price × 1000. This is not “AI performance per dollar”.
| # | GPU | VRAM | Current price | $ / advertised GB | tok/s per $1,000 | Winning SKU |
|---|---|---|---|---|---|---|
| 1 | NVIDIA RTX 3090 24 GB | 24 GB | $1347.22 Amazon | $56.1 | 34 · ESTIMATED · MEDIUM | GIGABYTE GIGABYTE EAGLE GV-N3090EAGLE OC-24GD graphics card NVIDIA GeForce RTX 3090 24 GB GDDR6X ASIN B08HJPDJTY |
| 2 | NVIDIA RTX 3090 Ti 24 GB | 24 GB | $2210.00 Amazon | $92.1 | 22 · ESTIMATED · MEDIUM | MSI MSI SUPRIM GeForce RTX 3090 Ti X 24G NVIDIA 24 GB GDDR6X ASIN B09Q31NH4M |
| 3 | NVIDIA RTX 4090 24 GB | 24 GB | $2579.99 Amazon | $107 | 21 · ESTIMATED · MEDIUM | GIGABYTE GIGABYTE AERO GeForce RTX 4090 OC 24G NVIDIA 24 GB GDDR6X ASIN B0BXMP9QVF |
| 4 | NVIDIA RTX 5090 32 GB | 32 GB | $4799.99 Amazon | $150 | 17.7 · ESTIMATED · MEDIUM | GIGABYTE GIGABYTE GeForce RTX 5090 WINDFORCE OC 32G Graphics Card ASIN B0DT7GMXHB |