Skip to content
Model Pareto

Nemotron 3 Super 120B A12B

NVIDIAOpen weightsartificialanalysis.ai ↗
Input
$0.30/M tok
Output
$0.90/M tok
Blended
$0.45/M tok · 3:1
Speed
164tok/s
Context
1Mtokens
Cost per task
~$0.38/taskest.

Long-tail import from Artificial Analysis (2026-09-24); AA variant: Nemotron 3 Super 120B A12B (Reasoning). Price = AA median host price.

Rankings

Self-hostingEstimated

How we estimate →

What it would cost to run these open weights yourself on rented GPUs, compared with the API price above.

Hardware
2× B200 180GB
FP8 weights
Throughput
~12,551 tok/s
many requests batched
Price
$0.11–0.96 /M tok
busy → light use
On your own machine
Mac Studio M5 Max 128GB (int4, ~72 tok/s single-stream)
single consumer GPU or Mac
Assumptions (5)
  • FP8 weights (120B params, 12B active per token) + 40% KV-cache headroom ≈ 168 GB
  • 2× B200 180GB at $5.98–$14.24/GPU-hour on-demand (2026-09-24)
  • ~12,551 output tok/s aggregate at batch 359 (bandwidth-bound); MoE compute scales with active params
  • Blended 3:1 input:output; prefill ~177,188 tok/s
  • Low = 75% utilization at the low GPU price; high = 20% at the high price

Benchmark scores

~ italic, dashed = no published score yet, estimated from related benchmarks.

Coding

Writing & chat

Hard reasoning

Agents

OSWorld 2.0Estimated
~last of 5 (estimated)below measured range (<47.5%)
  1. #1GLM-5.3 Flash59.1%
  2. #2Kimi K358.3%
  3. #4GPT-5.547.5%
  4. ~#5Nemotron 3 Super 120B A12BEstimated<47.5%
No published score — estimated from related benchmarks