Contributed byMario AlkaNVIDIA Nemotron-Cascade-2-30B-A3B
Performance benchmark · measured on 18.08.2026 07:39
Benchmark-ID
run-20260818-080548-cee82fTimebench 3 - Kombi (Prefill + Generation)MoE30BRuntime: llama.cppQuantisierung: IQ4_XS
Generation138,75tok/s
Prefill1.439,05tok/s
Time to First Token1.508,00ms
Total duration17,78s
Concurrency1parallel
Ranking in the field
9of 70 systems
Performance benchmark · Primary metric: Generation-Speed (tok/s) · 1× concurrent
This run is better than 88 % of all comparable systems.
Generation
138,8 tok/s
+161 % vs Ø 53,2
Prefill
1.439,1 tok/s
+15 % vs Ø 1.253,3
Time to First Token
1.508 ms
-86 % vs Ø 11.017
Distribution in the field1 – 276 tok/s
Wie schlägt sich dieser Benchmark mit anderen Modellen?
gemma-4-E2B-itNVIDIA GeForce RTX 5070 Ti · run-20260728-140953-194348
276,3 tok/s
gpt-oss-20bNVIDIA GeForce RTX 5070 Ti · run-20260727-145956-e481f0
242,3 tok/s
Nemotron-3-Nano-4BNVIDIA GeForce RTX 5070 Ti · run-20260727-090153-f59d6d
233,7 tok/s
gemma-4-E2B-itNVIDIA GeForce RTX 5070 Ti · run-20260727-032758-19b085
228,2 tok/s
Gemma-4-26B-A4B-itNVIDIA GeForce RTX 5070 Ti · run-20260729-032121-c5db85
189,1 tok/s
gemma-4-E4B-itNVIDIA GeForce RTX 5070 Ti · run-20260727-090152-82df54
165,0 tok/s
gemma-4-E4B-itNVIDIA GeForce RTX 5070 Ti · run-20260727-032758-68f60a
144,6 tok/s
Meta-Llama-3.1-8B-InstructNVIDIA GeForce RTX 5070 Ti · run-20260728-194135-5497cd
140,9 tok/s
Nemotron-Cascade-2-30B-A3B this runNVIDIA GeForce RTX 5070 Ti · run-20260818-080548-cee82f
138,8 tok/s
Ornith-1.0-9BNVIDIA GeForce RTX 5070 Ti · run-20260730-064527-5981fd
129,0 tok/s
Nemotron-Cascade-2-30B-A3B same modelNVIDIA GeForce RTX 5070 Ti · run-20260818-071253-cd7611
120,0 tok/s
gemma-4-12B-itNVIDIA GeForce RTX 5070 Ti · run-20260728-194136-e59252
92,8 tok/s
Laguna-XS-2.1NVIDIA GeForce RTX 5070 Ti · run-20260730-035053-523ecd
88,5 tok/s
Laguna-XS-2.1NVIDIA GeForce RTX 5070 Ti · run-20260729-224611-599fae
88,4 tok/s
How does this benchmark compare on other GPUs?
Same model on different hardware · 1× concurrent · Generation (tok/s)
NVIDIA GeForce RTX 5090run-20260729-105334-44d1e7
359,0 tok/s
NVIDIA RTX PRO 6000 Blackwell Workstation Editionrun-20260729-032120-4e3476
356,8 tok/s
3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Editionrun-20260729-105340-3ddb51
330,2 tok/s
3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Editionrun-20260729-105340-508364
329,9 tok/s
3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Editionrun-20260729-105340-f538c5
327,7 tok/s
NVIDIA GeForce RTX 3090 Tirun-20260729-060801-1e5cba
221,8 tok/s
NVIDIA GeForce RTX 5070 Ti this runrun-20260818-080548-cee82f
138,8 tok/s
NVIDIA GeForce RTX 5070 Tirun-20260818-071253-cd7611
120,0 tok/s
3× AMD Radeon AI PRO R9700run-20260722-181156-a273f2
107,9 tok/s
NVIDIA GB10run-20260820-124207-595ea8
77,0 tok/s
NVIDIA GeForce RTX 5070 Tirun-20260729-032125-03286f
69,0 tok/s
NVIDIA GB10run-20260820-124207-7e0db9
62,0 tok/s
3× AMD Radeon AI PRO R9700run-20260806-094441-85feb1
32,2 tok/s
3× AMD Radeon AI PRO R9700run-20260806-094441-502c90
31,8 tok/s
Hardware
GPU: NVIDIA GeForce RTX 5070 Ti · 16 GB VRAM
CPU: AMD Ryzen 9 7945HX with Radeon Graphics
RAM: 61 GB
Mainboard: Shenzhen Meigao Electronic Equipment Co.,Ltd DRFXI (MotherBoard Series)
Setup
Runtime: llama.cpp
Quantization: IQ4_XS
Model: Nemotron-Cascade-2-30B-A3B
Configuration
# LLM-Benchmark Konfiguration
# Modell : Nemotron-Cascade-2-30B-A3B
# Engine : llama.cpp
# Run-ID : run-20260818-080548-cee82f
# GPU : NVIDIA GeForce RTX 5070 Ti
# CPU : AMD Ryzen 9 7945HX with Radeon Graphics
# RAM : 61 GB
bench@llm-benchmark:~$ /root/llama.cpp/build-cuda/bin/llama-server \
-m /root/models/Nemotron-Cascade-2-30B-A3B-IQ4_XS.gguf \
-a Nemotron-Cascade-2-30B-A3B \
--host 0.0.0.0 \
--port 8000 \
-ngl 999 \
--n-cpu-moe 12 \
-c 40960 \
--parallel 10 \
-fa on \
--jinja
| Engine?Die Inferenz-Software, die das Modell ausliefert (z.B. vLLM oder llama.cpp). Sie bestimmt Geschwindigkeit, unterstuetzte Modellformate und welche Parameter ueberhaupt verfuegbar sind. | llamacpp |
| Modellalias?Der Name, unter dem das Modell ueber die API angesprochen wird. Genau dieser Wert muss im Request-Feld 'model' stehen. | Nemotron-Cascade-2-30B-A3B |
| Kontextlaenge?Maximale Anzahl Tokens (Eingabe + erzeugte Ausgabe zusammen), die das Modell pro Anfrage verarbeiten kann. | 40960 |
| Modellpfad?Pfad zur GGUF-Modelldatei, die geladen und ausgeliefert wird. | /root/models/Nemotron-Cascade-2-30B-A3B-IQ4_XS.gguf |
| Alias?Anzeigename des Modells nach aussen (served model name), unabhaengig vom Dateinamen. | Nemotron-Cascade-2-30B-A3B |
| GPU-Layer?Anzahl der auf die GPU ausgelagerten Modell-Layer. Hoeher = mehr VRAM und schneller; der Rest laeuft auf der CPU. 999 = alles auf GPU. | 999 |
| n-cpu-moe | 12 |
| Kontext?Groesse des Kontextfensters in Token. 0 = der beim Training verwendete Kontext des Modells. | 40960 |
| Parallel?Anzahl paralleler Slots/Sequenzen, die der Server gleichzeitig bedient. Der Kontext wird auf die Slots aufgeteilt. | 10 |
| fa | on |
Anzeige
Model comparison
Nemotron-Cascade-2-30B-A3B on various hardware
All published performance runs of this model – each bubble a variant: position = prefill (X) × generation (Y), bubble size = number of runs. Closer to the top right = faster. ★ Marked gold = this benchmark.
GPUby graphics card
NVIDIA RTX PRO 6000 Blackwell Workstation Edition 1.768,1 tok/sNVIDIA GeForce RTX 5090 1.719,2 tok/sNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition 1.655,2 tok/sNVIDIA GeForce RTX 3090 Ti 897,5 tok/sAMD Radeon 8060S Graphics 326,9 tok/s★ NVIDIA GeForce RTX 5070 Ti 262,4 tok/s this runNVIDIA GB10 (DGX Spark) 148,8 tok/sAMD Radeon AI PRO R9700 122,1 tok/sCPU-only 5,9 tok/s
CPUby processor
AMD Ryzen 9 9950X 16-Core Processor 1.768,1 tok/sAMD Ryzen 7 5800X3D 8-Core Processor 1.719,2 tok/sAMD Ryzen Threadripper PRO 9965WX 24-Cores 1.655,2 tok/sAMD Ryzen 9 8945HX with Radeon Graphics 897,5 tok/sAMD RYZEN AI MAX+ 395 w/ Radeon 8060S 326,9 tok/sAMD Ryzen Threadripper PRO 5975WX 32-Cores 262,4 tok/s★ AMD Ryzen 9 7945HX with Radeon Graphics 219,6 tok/s this runNVIDIA Grace 148,8 tok/sAMD Ryzen Threadripper PRO 7955WX 16-Cores 122,1 tok/sIntel(R) Xeon(R) CPU E5-4620 v2 @ 2.60GHz 5,9 tok/s
MBby mainboard
ASUSTeK COMPUTER INC. ProArt X870E-CREATOR WIFI 1.768,1 tok/sASUSTeK COMPUTER INC. ROG STRIX B550-A GAMING 1.719,2 tok/sASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE 1.655,2 tok/sMeigao Innovation Technology (Shen Zhen) Co., Ltd DRFXL (MotherBoard Series) 897,5 tok/sBosgame AXB35-02 (BeyondMax Series) 326,9 tok/sASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI 262,4 tok/s★ Shenzhen Meigao Electronic Equipment Co.,Ltd DRFXI (MotherBoard Series) 219,6 tok/s this runASUSTeK COMPUTER INC. GX10 148,8 tok/sDell Inc. PowerEdge R820 5,9 tok/s
DRVby driver
unbekannt 1.768,1 tok/sNVIDIA 590.48.01 / CUDA 13.1 148,8 tok/s
💰 Economics
Economics of this run
Operating cost, TCO and comparison with the next-best runs of the same model at identical concurrency (1× concurrent). Methodology →
⚙️ ConfigurationAll metrics and charts below follow these settings – based on a 24-month runtime.Save to URLReset
⚡ Electricity price
EUR/kWh
⚙️ System utilization
100 %
🖥️ Acquisition
EUR
🔌 Idle
0 W
⚡ TDP
300 W
☁️ External LLM (API)
Electricity0.30 EUR/kWh
Avg power (incl. idle)300 W estimated (TDP)GPU 300 W full load
Avg cost / hourEUR 0.090
Electricity / 1M tokensEUR 0.18
Token / kWh1.67M
Acquisition (system)EUR 120 missingPSU EUR 120
Electricity (2 years)–
TCO (2 years)EUR 1,697
Output tokens (2 years)8.75B
☁️ External LLM (API) – comparison
External LLM cost (2 years)–
Savings vs. external (2 years)–
All values above and the charts below take the configured system utilization into account: at X% the system generates only X% of the time, the rest it idles (0 W). Cost per hour drops (more idle), cost per token rises.
Cost over 2 years – electricity only
Cost over 2 years – incl. acquisition (TCO)
Speed vs. tokens per euro
Euro per 1M tokens
Comparison vs. API – economics per benchmark
| Nemotron-Cascade-2-30B-A3BNVIDIA GeForce RTX 5070 Ti | Nemotron-Cascade-2-30B-A3BNVIDIA GeForce RTX 5090 | Nemotron-Cascade-2-30B-A3BNVIDIA RTX PRO 6000 Blackwell Workstation Edition | Nemotron-Cascade-2-30B-A3B3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition | |
|---|---|---|---|---|
| Electricity cost (24 mo.) | – | – | – | – |
| Acquisition cost | – | – | – | – |
| Total cost (TCO) | – | – | – | – |
| Generated tokens (24 mo.) | – | – | – | – |
| Token price via API | – | – | – | – |
| Break-even point (days) | – | – | – | – |
| Result (savings / extra cost) | – | – | – | – |
NVIDIA GeForce RTX 5070 Ti · this run 139 tok/sNVIDIA GeForce RTX 5090 359 tok/sNVIDIA RTX PRO 6000 Blackwell Workstation Edition 357 tok/s3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition 330 tok/s
Comparison with up to 3 next-best runs of this model at the same concurrency (at least one on different hardware). Power = GPU TDP + CPU (idle + 15 %) + board (estimated), acquisition = full system (GPU + CPU + board + RAM + PSU), prices = stored market prices.
Contributed by
Mario Alka Administrator
@marioalka
Ich bin Unternehmer, Softwareentwickler und KI-Enthusiast. Seit vielen Jahren entwickle ich Unternehmenssoftware und beschäftige mich inzwischen fast täglich mit lokalen LLMs, KI-Agenten und leistungsfähiger KI-Hardware.
Mit LLM-Benchmark.de möchte ich eine Plattform schaffen, auf der Modelle, GPUs und Agenten objektiv und reproduzierbar miteinander verglichen werden.