# LLM-Benchmark Konfiguration # Modell : Qwen3-Omni-30B-A3B-Thinking # Engine : llama.cpp # Run-ID : run-20260831-132027-ba5d1c # GPU : AMD Radeon PRO W7800 48GB # CPU : AMD Ryzen Threadripper PRO 5975WX 32-Cores # RAM : 62 GB bench@llm-benchmark:~$ /home/godcore/llama.cpp/build_rocm/bin/llama-server \ -m Qwen3-Omni-30B-A3B-Thinking-Q4_K_M.gguf \ -ngl 999 \ -fa on \ -c 32768 \ -np 12 '(AMD' Radeon Pro W7800 48GB 'ROCm)'

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
Kontextlaenge?Maximale Anzahl Tokens (Eingabe + erzeugte Ausgabe zusammen), die das Modell pro Anfrage verarbeiten kann.	32768
Modellpfad?Pfad zur GGUF-Modelldatei, die geladen und ausgeliefert wird.	Qwen3-Omni-30B-A3B-Thinking-Q4_K_M.gguf
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
fa	on
Kontext?Groesse des Kontextfensters in Token. 0 = der beim Training verwendete Kontext des Modells.	32768
np	12

All benchmarks of this model To leaderboard

Model comparison

Qwen3-Omni-30B-A3B-Thinking 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 GeForce RTX 5090 1.673,6 tok/sNVIDIA RTX PRO 6000 Blackwell Workstation Edition 1.653,3 tok/sNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition 1.512,4 tok/sNVIDIA GeForce RTX 3090 Ti 927,4 tok/sNVIDIA RTX A6000 560,9 tok/s★ AMD Radeon PRO W7800 48GB 205,1 tok/s this runAMD Radeon AI PRO R9700 160,6 tok/sNVIDIA GeForce RTX 5070 Ti 155,9 tok/s

CPUby processor

AMD Ryzen 7 5800X3D 8-Core Processor 1.673,6 tok/sAMD Ryzen 9 9950X 16-Core Processor 1.653,3 tok/sAMD Ryzen Threadripper PRO 9965WX 24-Cores 1.512,4 tok/sAMD Ryzen 9 8945HX with Radeon Graphics 927,4 tok/sAMD Ryzen Threadripper PRO 7955WX 16-Cores 560,9 tok/s★ AMD Ryzen Threadripper PRO 5975WX 32-Cores 205,1 tok/s this run

MBby mainboard

ASUSTeK COMPUTER INC. ROG STRIX B550-A GAMING 1.673,6 tok/sASUSTeK COMPUTER INC. ProArt X870E-CREATOR WIFI 1.653,3 tok/sASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE 1.512,4 tok/sMeigao Innovation Technology (Shen Zhen) Co., Ltd DRFXL (MotherBoard Series) 927,4 tok/s★ ASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI 205,1 tok/s this run

ENGby engine

★ llama.cpp 1.673,6 tok/s this runvLLM 910,7 tok/sunbekannt 34,8 tok/s

DRVby driver

unbekannt 1.673,6 tok/sAMD 7.0.0-27-generic 34,8 tok/s

💰 Economics

Economics of this run

Operating cost, TCO and comparison with the next-best runs of the same model at identical concurrency (10× concurrent). Methodology →

⚙️ ConfigurationAll metrics and charts below follow these settings – based on a 24-month runtime.Save to URL Reset

⚡ Electricity price EUR/kWh

⚙️ System utilization 100 %

🖥️ Acquisition EUR

🔌 Idle 10 W

⚡ TDP 270 W

☁️ External LLM (API) show

Input €/1M Output €/1M Input per output token

Electricity0.30 EUR/kWh

Avg power (incl. idle)270 W estimated (TDP)GPU 260 + Board 10 W full load

Avg cost / hourEUR 0.081

Electricity / 1M tokensEUR 0.11

Token / kWh2.73M

Acquisition (system)EUR 5,304 partial priceGPU EUR 2,808 · CPU EUR 1,880 · RAM EUR 496 · PSU EUR 120

Electricity (2 years)–

TCO (2 years)EUR 6,723

Output tokens (2 years)12.94B

☁️ 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 (10 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

	Qwen3-Omni-30B-A3B-ThinkingAMD Radeon PRO W7800 48GB	Qwen3-Omni-30B-A3B-ThinkingNVIDIA GeForce RTX 5090	Qwen3-Omni-30B-A3B-ThinkingNVIDIA RTX PRO 6000 Blackwell Workstation Edition	Qwen3-Omni-30B-A3B-Thinking3x 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)	–	–	–	–

AMD Radeon PRO W7800 48GB · this run 205 tok/s NVIDIA GeForce RTX 5090 1,674 tok/s NVIDIA RTX PRO 6000 Blackwell Workstation Edition 1,653 tok/s 3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition 1,512 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.