# LLM-Benchmark Konfiguration # Modell : Qwen2.5-32B-Instruct-AWQ # Engine : llama.cpp # Run-ID : run-20260809-131004-1cb07c # GPU : NVIDIA GeForce RTX 3090 Ti # CPU : AMD Ryzen 9 8945HX with Radeon Graphics # RAM : 92 GB bench@llm-benchmark:~$ /root/llama.cpp/build/bin/llama-server \ -m /root/.cache/huggingface/hub/models--bartowski--Qwen2.5-32B-Instruct-GGUF/snapshots/2116cbb385b8ce3a4d28cf3bf1cd2039a55821a6/Qwen2.5-32B-Instruct-Q4_K_M.gguf \ --alias Qwen2.5-32B-Instruct-AWQ \ --host 0.0.0.0 \ --port 8000 \ -ngl 999 \ -c 16384 \ -np 4 \ --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.	Qwen2.5-32B-Instruct-AWQ
Kontextlaenge?Maximale Anzahl Tokens (Eingabe + erzeugte Ausgabe zusammen), die das Modell pro Anfrage verarbeiten kann.	16384
Modellpfad?Pfad zur GGUF-Modelldatei, die geladen und ausgeliefert wird.	/root/.cache/huggingface/hub/models--bartowski--Qwen2.5-32B-Instruct-GGUF/snapshots/2116cbb385b8ce3a4d28cf3bf1cd2039a55821a6/Qwen2.5-32B-Instruct-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
Kontext?Groesse des Kontextfensters in Token. 0 = der beim Training verwendete Kontext des Modells.	16384
np	4
execution_type	local

All benchmarks of this model To leaderboard

Model comparison

Qwen2.5-32B-Instruct-AWQ 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 510,3 tok/sNVIDIA GeForce RTX 5090 473,9 tok/sNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition 412,6 tok/sNVIDIA RTX A6000 362,6 tok/s★ NVIDIA GeForce RTX 3090 Ti 223,8 tok/s this runAMD Radeon RX 7900 XTX 179,0 tok/sAMD Radeon AI PRO R9700 116,3 tok/sAMD Radeon PRO W7900 Dual Slot 89,6 tok/sAMD Radeon PRO W7800 48GB 69,9 tok/sNVIDIA GeForce RTX 5070 Ti 45,0 tok/s

CPUby processor

AMD Ryzen 9 9950X 16-Core Processor 510,3 tok/sAMD Ryzen 7 5800X3D 8-Core Processor 473,9 tok/sAMD Ryzen Threadripper PRO 9965WX 24-Cores 412,6 tok/sAMD Ryzen Threadripper PRO 7955WX 16-Cores 362,6 tok/s★ AMD Ryzen 9 8945HX with Radeon Graphics 223,8 tok/s this runAMD Ryzen Threadripper PRO 3955WX 16-Cores 179,0 tok/sAMD Ryzen 3 3100 4-Core Processor 116,3 tok/sAMD Ryzen Threadripper PRO 5975WX 32-Cores 89,6 tok/s

MBby mainboard

ASUSTeK COMPUTER INC. ProArt X870E-CREATOR WIFI 510,3 tok/sASUSTeK COMPUTER INC. ROG STRIX B550-A GAMING 473,9 tok/sASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE 412,6 tok/s★ Meigao Innovation Technology (Shen Zhen) Co., Ltd DRFXL (MotherBoard Series) 223,8 tok/s this runASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI 179,0 tok/sASUSTeK COMPUTER INC. ROG STRIX X570-F GAMING 116,3 tok/s

ENGby engine

vLLM 510,3 tok/s★ llama.cpp 474,0 tok/s this run

DRVby driver

unbekannt 510,3 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 50 W

⚡ TDP 477 W

☁️ External LLM (API) show

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

Electricity0.30 EUR/kWh

Avg power (incl. idle)477 W estimated (TDP)GPU 450 + CPU 17 + Board 10 W full load

Avg cost / hourEUR 0.14

Electricity / 1M tokensEUR 0.18

Token / kWh1.67M

Acquisition (system)EUR 2,986 partial priceGPU EUR 999 · CPU EUR 549 · RAM EUR 1,288 · PSU EUR 150

Electricity (2 years)–

TCO (2 years)EUR 5,494

Output tokens (2 years)14.00B

☁️ 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 (50 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

	Qwen2.5-32B-Instruct-AWQNVIDIA GeForce RTX 3090 Ti	Qwen2.5-32B-Instruct-AWQNVIDIA RTX PRO 6000 Blackwell Workstation Edition	Qwen2.5-32B-Instruct-AWQNVIDIA RTX PRO 6000 Blackwell Workstation Edition	Qwen2.5-32B-Instruct-AWQNVIDIA GeForce RTX 5090
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 3090 Ti · this run 222 tok/s NVIDIA RTX PRO 6000 Blackwell Workstation Edition 510 tok/s NVIDIA RTX PRO 6000 Blackwell Workstation Edition 474 tok/s NVIDIA GeForce RTX 5090 474 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.