Skip to content
Model Pareto

Nemotron 3.5 Lightning

NVIDIAOpen weightsartificialanalysis.ai ↗
Input
$0.06/M tok
Output
$0.20/M tok
Blended
$0.095/M tok · 3:1
Speed
267tok/s
Context
1Mtokens
Cost per task
~$0.08/taskest.

Long-tail import from Artificial Analysis (2026-09-24); AA variant: Nemotron 3.5 Lightning. 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
1× H200 141GB
FP8 weights
Throughput
~8,799 tok/s
many requests batched
Price
$0.044–0.37 /M tok
busy → light use
On your own machine
RTX 5090 32GB (int4, 300+ tok/s single-stream)
single consumer GPU or Mac
Assumptions (5)
  • FP8 weights (30B params, 3B active per token) + 40% KV-cache headroom ≈ 42 GB
  • 1× H200 141GB at $3.59–$7.91/GPU-hour on-demand (2026-09-24)
  • ~8,799 output tok/s aggregate at batch 256 (bandwidth-bound); MoE compute scales with active params
  • Blended 3:1 input:output; prefill ~155,846 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

~#25 of 26 (estimated)~28.0
  1. #1Claude Opus 5.561.7
  2. #2Claude Opus 559.7
  3. #24GPT-6 Luna31.1
  4. ~#25Nemotron 3.5 LightningEstimated28.0
  5. #25Muse Spark Open26.3
No published score — estimated from related benchmarks

Hard reasoning

Agents

~#12 of 15 (estimated)~59.7%
  1. #1Kimi K384.8%
  2. #2Qwen3.8 27B84.3%
  3. #11Muse Glimmer65.9%
  4. ~#12Nemotron 3.5 LightningEstimated59.7%
  5. #12Nemotron 3.5 Super VL NVFP452.3%
No published score — estimated from related benchmarks
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 3.5 LightningEstimated<47.5%
No published score — estimated from related benchmarks