Best hardware for Nemotron 3 Super 120B-A12B
Deep reasoning and long-form answers.
121B parameters · MoE (12B active per token, so it runs faster than its size suggests) · nemotron family
Our picks for this model
Cheapest · Easiest · Fastest
$6,050 on eBay ↗ · ~54.3 tok/s est.
multi-card build (needs a bigger PSU and board)
No single card has enough VRAM for this model, so this is the cheapest clean multi-card build.
Rule-based picks at Q4_K_M, from live screened prices (how we estimate speed and choose picks). Every option is ranked below.
Q4_K_M
Minimum VRAM: 93.33 GB (file 86.98 GB + context/runtime headroom at 8k ctx; see methodology).
No tracked cloud offer fits this quant's VRAM needs, so there is no breakeven to compute; any fitting hardware is shown buy-only.
| Config | Screened price | Total VRAM | Est. watts | Est. tokens/sec | Breakeven vs cloud | Verdict |
|---|---|---|---|---|---|---|
| 4× NVIDIA GeForce RTX 3090 our pick · Cheapest · Easiest · Fastest | $6,050 | 96 GB | 1475 W | ~54.3 | – | Buy-only; no cloud offer fits |
| 2× NVIDIA RTX A6000 | $8,570 | 96 GB | 675 W | ~44.5 | – | Buy-only; no cloud offer fits |
| 4× NVIDIA Tesla P40 Passive datacenter card: needs a cooling mod and is far slower than modern cards | $1,256 | 96 GB | 1075 W | ~20.1 (well below this table's average) | – | Buy-only; no cloud offer fits |
Fits, but not currently buyable as a set: 4× NVIDIA RTX A5000 (96 GB) needs 4 screened cards and the market has 2. We price a multi-card build from that many separate listings, so it stays unpriced until they exist.
- Our pick rows sit on top, then clean configs by price; cards with a caveat (cooling mod needed, or poor value for AI) group last, whatever their price. Watts and tokens/sec turn green or red when they sit far from this table's average (roughly 40% either way).
- Prices link straight to the screened eBay listing; hover for the seller summary. Seller details and the report option live on each card's own page.
- Configs without a screened purchase option are hidden until listings clear the trust screen (a multi-day survival window on eBay).
- Multi-card costs are the sum of that many separate screened listings; the cheapest listing counted twice is not a price anyone can pay. Prices exclude the PSU, board and case a multi-card build usually also needs.
- Tokens/sec are estimates, not benchmarks: spec-sheet memory bandwidth at 50% utilization divided by the model bytes read per token (how and why). Real speeds vary with the runtime and context length.
- Breakeven and verdict use default assumptions (40 h/mo, $0.16/kWh, 24-month horizon, resale at 65% of the going rate); adjust them in the calculator. Hardware links are screened, not guaranteed.
Run your own numbers for Nemotron 3 Super 120B-A12B (Q4_K_M) in the buy-vs-rent calculator →
Q8_0
Minimum VRAM: 136.9 GB (file 128.47 GB + context/runtime headroom at 8k ctx; see methodology).
No tracked cloud offer fits this quant's VRAM needs, so there is no breakeven to compute; any fitting hardware is shown buy-only.
No screened hardware configs currently fit this quant.
- Our pick rows sit on top, then clean configs by price; cards with a caveat (cooling mod needed, or poor value for AI) group last, whatever their price. Watts and tokens/sec turn green or red when they sit far from this table's average (roughly 40% either way).
- Prices link straight to the screened eBay listing; hover for the seller summary. Seller details and the report option live on each card's own page.
- Configs without a screened purchase option are hidden until listings clear the trust screen (a multi-day survival window on eBay).
- Multi-card costs are the sum of that many separate screened listings; the cheapest listing counted twice is not a price anyone can pay. Prices exclude the PSU, board and case a multi-card build usually also needs.
- Tokens/sec are estimates, not benchmarks: spec-sheet memory bandwidth at 50% utilization divided by the model bytes read per token (how and why). Real speeds vary with the runtime and context length.
- Breakeven and verdict use default assumptions (40 h/mo, $0.16/kWh, 24-month horizon, resale at 65% of the going rate); adjust them in the calculator. Hardware links are screened, not guaranteed.
Run your own numbers for Nemotron 3 Super 120B-A12B (Q8_0) in the buy-vs-rent calculator →