Contributed byMario AlkaQwen (Alibaba) Qwen3-VL-30B-A3B-Instruct
Performance benchmark · measured on 08.10.2026 11:41
Benchmark-ID
run-20261009-081230-0ec464Timebench 3 - Kombi (Prefill + Generation)MoE30BRuntime: llama.cppQuantisierung: Q4_K_M
ℹ
Public note: Dell PowerEdge R820 . 2x AMD Radeon RX 7900 XTX (48 GB VRAM gesamt, ROCm/HIP) . RAM: 32x 16 GB SK Hynix DDR3-1600 ECC RDIMM (HMT42GR7AFR4A-PB) = 512 GB @ 1600 MT/s
Generation221,62tok/s
Prefill6.455,25tok/s
Time to First Token4.958,00ms
Total duration104,02s
Concurrency10parallel
Ranking in the field
775of 1427 systems
Performance benchmark · Primary metric: Generation-Speed (tok/s) · 10× concurrent
This run is better than 46 % of all comparable systems.
Generation
221,6 tok/s
-48 % vs Ø 425,3
Prefill
6.455,3 tok/s
-3 % vs Ø 6.659,3
Time to First Token
4.958 ms
-89 % vs Ø 45.848
Distribution in the field0 – 2.491 tok/s
Wie schlägt sich dieser Benchmark mit anderen Modellen?
gemma-4-E2B-itNVIDIA GeForce RTX 5090 · run-20260728-184455-e28775
2.491,2 tok/s
gemma-4-E2B-itNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260727-181606-b37ed9
2.414,4 tok/s
Nemotron-3.5-Lightning-30B-A3BNVIDIA GeForce RTX 3090 Ti · run-20261021-093815-dce558
2.263,7 tok/s
Nemotron-3-Nano-4BNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260727-181607-4cca42
2.182,7 tok/s
gemma-4-E2B-it3× 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-184456-d87c73
2.143,6 tok/s
gpt-oss-20bNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260730-035052-9c966e
1.995,4 tok/s
gpt-oss-20bNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260728-140954-8007c4
1.993,5 tok/s
Nemotron-3-Nano-4B3× 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-194136-0aed86
1.937,7 tok/s
Nemotron-3-Nano-4B3× 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-194136-8db0fa
1.911,6 tok/s
Nemotron-3-Nano-4B3× 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260727-181606-e0eafd
1.897,0 tok/s
gpt-oss-20bNVIDIA GeForce RTX 5090 · run-20260729-032121-7c961d
1.890,7 tok/s
Nemotron-3-Nano-4BNVIDIA GeForce RTX 5090 · run-20260728-194135-2ed9dd
1.878,7 tok/s
gpt-oss-20b3× 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-034052-26063b
1.835,7 tok/s
Nemotron-3-Nano-4B3× 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-194136-a77046
1.830,3 tok/s
Qwen3-VL-30B-A3B-Instruct this run2× 2x AMD Radeon RX 7900 XTX · run-20261009-081230-0ec464
221,6 tok/s
How does this benchmark compare on other GPUs?
Same model on different hardware · 10× concurrent · Generation (tok/s)
NVIDIA GeForce RTX 5090run-20260729-105336-473379
1.655,9 tok/s
NVIDIA RTX PRO 6000 Blackwell Workstation Editionrun-20260729-032125-329bcc
1.639,0 tok/s
3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Editionrun-20260729-122848-b544db
1.505,0 tok/s
3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Editionrun-20260729-122848-e355fe
1.498,9 tok/s
3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Editionrun-20260729-105343-cd178b
1.420,2 tok/s
NVIDIA RTX PRO 6000 Blackwell Workstation Editionrun-20260724-004607-c49303
1.028,5 tok/s
NVIDIA GeForce RTX 3090 Tirun-20260729-105331-2e293f
903,5 tok/s
2× NVIDIA RTX A6000run-20260829-124021-34de42
763,5 tok/s
2× NVIDIA RTX A6000run-20260827-090902-dce1be
697,0 tok/s
2× NVIDIA RTX A6000run-20260829-191617-b93726
547,7 tok/s
2× NVIDIA RTX A6000run-20260827-211025-7e41cd
540,4 tok/s
2× NVIDIA RTX A6000run-20260830-040426-7a8075
536,3 tok/s
2× NVIDIA RTX A6000run-20260827-140321-436bf6
516,4 tok/s
NVIDIA RTX A6000run-20260826-211812-d8d46f
440,0 tok/s
Hardware
GPU: 2x AMD Radeon RX 7900 XTX · 24 GB VRAM
CPU: 4x Intel(R) Xeon(R) CPU E5-4657L v2 @ 2.40GHz
RAM: 504 GB
Mainboard: Dell Inc. PowerEdge R820
Setup
Runtime: llama.cpp
Quantization: Q4_K_M
Model: Qwen3-VL-30B-A3B-Instruct
Configuration
# LLM-Benchmark Konfiguration
# Modell : Qwen3-VL-30B-A3B-Instruct
# Engine : llama.cpp
# Run-ID : run-20261009-081230-0ec464
# GPU : 2x AMD Radeon RX 7900 XTX
# CPU : 4x Intel(R) Xeon(R) CPU E5-4657L v2 @ 2.40GHz
# RAM : 504 GB
bench@llm-benchmark:~$ llama-server \
-m Qwen_Qwen3-VL-30B-A3B-Instruct-IQ4_XS.gguf \
--alias Qwen3-VL-30B-A3B-Instruct \
-ngl 999 \
-sm layer \
-fa on \
-c 40960 \
-np 10 '(2x' RX 7900 XTX ROCm/HIP, 'gpu)'
| 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. | Qwen3-VL-30B-A3B-Instruct |
| 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. | Qwen_Qwen3-VL-30B-A3B-Instruct-IQ4_XS.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 |
| Split-Mode?Verteilung ueber mehrere GPUs: none (nur eine GPU), layer (Layer aufteilen) oder row (Tensoren zeilenweise). | layer |
| fa | on |
| Kontext?Groesse des Kontextfensters in Token. 0 = der beim Training verwendete Kontext des Modells. | 40960 |
| np | 10 |
| execution_type | local |
Anzeige
Model comparison
Qwen3-VL-30B-A3B-Instruct 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 1.655,9 tok/sNVIDIA RTX PRO 6000 Blackwell Workstation Edition 1.639,0 tok/sNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition 1.505,0 tok/sNVIDIA GeForce RTX 3090 Ti 903,5 tok/sNVIDIA RTX A6000 763,5 tok/s★ AMD Radeon RX 7900 XTX 301,6 tok/s this runAMD Radeon PRO W7900 Dual Slot 245,1 tok/sAMD Radeon AI PRO R9700 233,2 tok/sAMD Radeon PRO W7800 48GB 233,0 tok/sNVIDIA GeForce RTX 5070 Ti 154,0 tok/s
CPUby processor
AMD Ryzen 7 5800X3D 8-Core Processor 1.655,9 tok/sAMD Ryzen 9 9950X 16-Core Processor 1.639,0 tok/sAMD Ryzen Threadripper PRO 9965WX 24-Cores 1.505,0 tok/sAMD Ryzen 9 8945HX with Radeon Graphics 903,5 tok/sAMD Ryzen Threadripper PRO 7955WX 16-Cores 763,5 tok/sAMD Ryzen Threadripper PRO 3955WX 16-Cores 301,6 tok/sAMD Ryzen Threadripper PRO 5975WX 32-Cores 245,1 tok/sAMD Ryzen 3 3100 4-Core Processor 233,2 tok/s★ Intel(R) Xeon(R) CPU E5-4657L v2 @ 2.40GHz 221,6 tok/s this run
MBby mainboard
ASUSTeK COMPUTER INC. ROG STRIX B550-A GAMING 1.655,9 tok/sASUSTeK COMPUTER INC. ProArt X870E-CREATOR WIFI 1.639,0 tok/sASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE 1.505,0 tok/sMeigao Innovation Technology (Shen Zhen) Co., Ltd DRFXL (MotherBoard Series) 903,5 tok/sASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI 301,6 tok/sASUSTeK COMPUTER INC. ROG STRIX X570-F GAMING 233,2 tok/s★ Dell Inc. PowerEdge R820 221,6 tok/s this run
DRVby driver
unbekannt 1.655,9 tok/sAMD 7.0.0-27-generic 89,8 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 URLReset
⚡ Electricity price
EUR/kWh
⚙️ System utilization
100 %
🖥️ Acquisition
EUR
🔌 Idle
35 W
⚡ TDP
749 W
☁️ External LLM (API)
Electricity0.30 EUR/kWh
Avg power (incl. idle)749 W estimated (TDP)GPU 710 + CPU 29 + Board 10 W full load
Avg cost / hourEUR 0.22
Electricity / 1M tokensEUR 0.28
Token / kWh1.07M
Acquisition (system)EUR 4,353 partial priceGPU EUR 2,098 · CPU EUR 59 · RAM EUR 2,016 · PSU EUR 180
Electricity (2 years)–
TCO (2 years)EUR 8,288
Output tokens (2 years)13.98B
☁️ 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 (35 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-VL-30B-A3B-Instruct2x AMD Radeon RX 7900 XTX | Qwen3-VL-30B-A3B-InstructNVIDIA GeForce RTX 5090 | Qwen3-VL-30B-A3B-InstructNVIDIA RTX PRO 6000 Blackwell Workstation Edition | Qwen3-VL-30B-A3B-Instruct3x 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) | – | – | – | – |
2× AMD Radeon RX 7900 XTX · this run 222 tok/sNVIDIA GeForce RTX 5090 1,656 tok/sNVIDIA RTX PRO 6000 Blackwell Workstation Edition 1,639 tok/s3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition 1,505 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.