Nemotron Cascade 2 30B A3B
- Input
- —
- Output
- —
- Speed
- ~200tok/sest.
- Context
- 1Mtokens
Long-tail import from Artificial Analysis (2026-09-24); AA variant: Nemotron Cascade 2 30B A3B. Price = AA median host price.
Rankings
Self-hostingEstimated
How we estimate →What it would cost to run these open weights yourself on rented GPUs. No API sells this model, so this is its price on the chart.
- 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.
Writing & chat
IFBenchVerified
#8 of 13980.4%
- #1MiniMax-M382.9%
- #2Qwen3.8 2.4T A95B82.8%
- #7Intern-S2-Preview (35B-A3B)80.9%
- #8Nemotron Cascade 2 30B A3B80.4%
- #9Inkling79.8%
artificialanalysis.ai · 2026-09-24
MMMLUEstimated
~#3 of 5 (estimated)~86.8%
- #1Gemini 3.1 Pro92.6%
- #2Gemma 4 31B88.4%
- ~#3Nemotron Cascade 2 30B A3BEstimated86.8%
- #3Claude Haiku 4.583.0%
No published score — estimated from related benchmarks
LMArena Text, Non-English (Elo)Estimated
~#42 of 57 (estimated)~1395 Elo
- #1Gemini 4 ArgonFew votes1512
- #2Claude Opus 5.51497
- #41Claude Haiku 4.51395
- ~#42Nemotron Cascade 2 30B A3BEstimated1395
- #42GPT-5.4 nano1387
No published score — estimated from related benchmarks
LMArena Text (Elo)Estimated
~#42 of 57 (estimated)~1411 Elo
- #1Gemini 4 Argon1525
- #2Claude Opus 5.51507
- #41Mistral Large 31413
- ~#42Nemotron Cascade 2 30B A3BEstimated1411
- #42GPT-5.4 nano1402
No published score — estimated from related benchmarks
EQ-Bench Creative Writing v3 (Elo)Estimated
~#18 of 26 (estimated)~1634 Elo
- #1GPT-6 Astra2173
- #2Claude Fable 5.12162
- #17Qwen3.8 27B1671
- ~#18Nemotron Cascade 2 30B A3BEstimated1634
- #18Inkling1611
No published score — estimated from related benchmarks
Hard reasoning
GPQA DiamondVerified
#90 of 19275.8%
- #1GPT-6 Astra96.1%
- #2Gemini 3.8 Flash95.3%
- #89Nova 2.0 Omni76.0%
- #90Nemotron Cascade 2 30B A3B75.8%
- #91North Mini Code75.7%
artificialanalysis.ai · 2026-09-24
Humanity's Last ExamVerified
#98 of 19812.0%
- #1Claude Fable 5.274.3%
- #2Claude Opus 5.561.4%
- #98Command A+12.0%
- #98Nemotron Cascade 2 30B A3B12.0%
- #101Mercury 2.511.8%
artificialanalysis.ai · 2026-09-24
AIME (latest)Estimated
~#23 of 34 (estimated)~72.1%
- #1GLM-5.3 NVFP499.5%
- #2Beam 1.0 501B A23B97.7%
- #22Phi-4-reasoning-plus78.0%
- ~#23Nemotron Cascade 2 30B A3BEstimated72.1%
- #23Mistral Large 338.0%
No published score — estimated from related benchmarks
AA Intelligence IndexEstimated
~#70 of 103 (estimated)~16.6
- #1Gemini 4 Carbon59.4
- #2Claude Opus 5.557.6
- #69Ring-2.6-1T16.6
- ~#70Nemotron Cascade 2 30B A3BEstimated16.6
- #70K2 Horizon 3.7B15.6
No published score — estimated from related benchmarks
Agents
τ²-bench (Telecom)Verified
#50 of 11953.2%
- #1Step 3.7 Flash98.5%
- #2DeepSeek-V4-Pro-0813-nvfp4-DSpark98.3%
- #49Claude Haiku 4.554.7%
- #50Nemotron Cascade 2 30B A3B53.2%
- #51Solar Open 100B48.2%
artificialanalysis.ai · 2026-09-24
OSWorld-VerifiedEstimated
~#12 of 15 (estimated)~61.7%
- #1Kimi K384.8%
- #2Qwen3.8 27B84.3%
- #11Muse Glimmer65.9%
- ~#12Nemotron Cascade 2 30B A3BEstimated61.7%
- #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%)
No published score — estimated from related benchmarks
BrowseCompEstimated
~#19 of 22 (estimated)~58.9%
- #1Atria Dawn Preview92.5%
- #2GPT-6 Astra91.5%
- #18Nemotron 3.5 Bolt 74B-A9B70.0%
- ~#19Nemotron Cascade 2 30B A3BEstimated58.9%
- #19Mistral Medium 3.548.6%
No published score — estimated from related benchmarks
GDPval-AA (Elo)Verified
#85 of 91252 Elo
- #1Claude Opus 5.51866
- #2Muse Wonder Full1857
- #84Celeris-1288
- #85Nemotron Cascade 2 30B A3B252
- #86Ministral 3 14B234
artificialanalysis.ai · 2026-09-24
AutomationBench-AAEstimated
~#46 of 85 (estimated)~15.3%
- #1Claude Fable 5.276.4%
- #2Muse Wonder Full71.3%
- #45Ling-3.0-flash-VL15.7%
- ~#46Nemotron Cascade 2 30B A3BEstimated15.3%
- #46K2 Horizon 3.7B14.7%
No published score — estimated from related benchmarks
τ-Bench Banking (AA)Estimated
~#47 of 91 (estimated)~15.6%
- #1Muse Spark 1.350.5%
- #2GLM-5.350.3%
- #46A.X-K216.1%
- ~#47Nemotron Cascade 2 30B A3BEstimated15.6%
- #47MiniMax-M315.3%
No published score — estimated from related benchmarks