Contributed byMario AlkaQwen (Alibaba) Qwen3-30B-A3B-Thinking-2507
Performance benchmark · measured on 24.10.2026 10:37
Benchmark-ID
run-20261024-084507-0f8d84Timebench 3 - Kombi (Prefill + Generation)MoE30BRuntime: llama.cppQuantisierung: Q4_K_M
ℹ
Public note: NVIDIA GeForce RTX 3090 24GB . Qwen3-30B-A3B-Thinking-2507 Q4_K_M . llama.cpp server-cuda (b11804), -ngl 999, -np 12, -c 49152 (4096 Tok/Slot)
Generation371,66tok/s
Prefill8.744,79tok/s
Time to First Token3.710,00ms
Total duration64,57s
Concurrency10parallel
Ranking in the field
4of 6 systems
Performance benchmark · Primary metric: Generation-Speed (tok/s) · 10× concurrent
This run is better than 40 % of all comparable systems.
Generation
371,7 tok/s
-27 % vs Ø 512,1
Prefill
8.744,8 tok/s
+126 % vs Ø 3.865,8
Time to First Token
3.710 ms
-90 % vs Ø 36.661
Distribution in the field140 – 1.133 tok/s
Wie schlägt sich dieser Benchmark mit anderen Modellen?
Qwen3.8-Flash-NextNVIDIA GeForce RTX 3090 · run-20261013-092430-0e9347
1.132,7 tok/s
Qwen3.8-Flash-NextNVIDIA GeForce RTX 3090 · run-20261013-165111-37735f
638,6 tok/s
gpt-oss-20bNVIDIA GeForce RTX 3090 · run-20261024-084507-9d9121
448,2 tok/s
Qwen3-30B-A3B-Thinking-2507 this runNVIDIA GeForce RTX 3090 · run-20261024-084507-0f8d84
371,7 tok/s
Qwen3.6-35B-A3BNVIDIA GeForce RTX 3090 · run-20261024-084508-cb8295
341,5 tok/s
Nemotron-3.5-Lightning-30B-A3BNVIDIA GeForce RTX 3090 · run-20261024-090453-f3a9ca
140,0 tok/s
How does this benchmark compare on other GPUs?
Same model on different hardware · 10× concurrent · Generation (tok/s)
NVIDIA GeForce RTX 5090run-20260729-105335-e6f72a
1.683,8 tok/s
NVIDIA RTX PRO 6000 Blackwell Workstation Editionrun-20260729-032123-88fa84
1.670,5 tok/s
NVIDIA RTX PRO 6000 Blackwell Workstation Editionrun-20261023-065850-7d5629
1.623,2 tok/s
3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Editionrun-20260729-105342-9f1be6
1.521,4 tok/s
3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Editionrun-20260729-105342-a89e7f
1.518,4 tok/s
3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Editionrun-20260729-105342-bd5e1f
1.465,3 tok/s
NVIDIA GeForce RTX 3090 Tirun-20260729-060803-f260ab
930,6 tok/s
NVIDIA RTX PRO 6000 Blackwell Workstation Editionrun-20261023-071954-1cd2ee
913,3 tok/s
NVIDIA RTX PRO 6000 Blackwell Workstation Editionrun-20260724-004605-f9a3a1
906,0 tok/s
NVIDIA GeForce RTX 3090 Tirun-20260724-004603-ee41df
883,7 tok/s
NVIDIA GeForce RTX 5090run-20260723-045450-14f178
873,7 tok/s
2× NVIDIA RTX A6000run-20260829-124021-f57a00
751,7 tok/s
2× NVIDIA RTX A6000run-20260827-090902-07d874
673,2 tok/s
2× NVIDIA RTX A6000run-20260829-181042-e10c46
554,2 tok/s
Hardware
GPU: NVIDIA GeForce RTX 3090 · 24 GB VRAM
CPU: AMD Ryzen 5 5600X 6-Core Processor
RAM: 31 GB
Mainboard: ASUSTeK COMPUTER INC. ROG STRIX B550-A GAMING
Setup
Runtime: llama.cpp
Quantization: Q4_K_M
Model: Qwen3-30B-A3B-Thinking-2507
Configuration
# LLM-Benchmark Konfiguration
# Modell : Qwen3-30B-A3B-Thinking-2507
# Engine : llama.cpp
# Run-ID : run-20261024-084507-0f8d84
# GPU : NVIDIA GeForce RTX 3090
# CPU : AMD Ryzen 5 5600X 6-Core Processor
# RAM : 31 GB
bench@llm-benchmark:~$ llama-server \
-m /models/Qwen_Qwen3-30B-A3B-Thinking-2507-Q4_K_M.gguf \
-ngl 999 \
-np 12 \
-c 49152
| 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. | Qwen3-30B-A3B-Thinking-2507 |
| Kontextlaenge?Maximale Anzahl Tokens (Eingabe + erzeugte Ausgabe zusammen), die das Modell pro Anfrage verarbeiten kann. | 49152 |
| Modellpfad?Pfad zur GGUF-Modelldatei, die geladen und ausgeliefert wird. | /models/Qwen_Qwen3-30B-A3B-Thinking-2507-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 |
| np | 12 |
| Kontext?Groesse des Kontextfensters in Token. 0 = der beim Training verwendete Kontext des Modells. | 49152 |
| execution_type | local |
Anzeige
Model comparison
Qwen3-30B-A3B-Thinking-2507 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 1.683,8 tok/sNVIDIA RTX PRO 6000 Blackwell Workstation Edition 1.670,5 tok/sNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition 1.521,4 tok/sNVIDIA GeForce RTX 3090 Ti 930,6 tok/sNVIDIA RTX A6000 751,7 tok/s★ NVIDIA GeForce RTX 3090 371,7 tok/s this runAMD Radeon RX 7900 XTX 303,1 tok/sAMD Radeon PRO W7800 48GB 258,3 tok/sAMD Radeon PRO W7900 Dual Slot 254,6 tok/sAMD Radeon AI PRO R9700 238,8 tok/sNVIDIA DGX SPARK (GB10) 161,1 tok/sNVIDIA GeForce RTX 5070 Ti 155,8 tok/sNVIDIA RTX SPARK (N1X) 107,0 tok/sCPU-only 8,5 tok/s
CPUby processor
AMD Ryzen 7 5800X3D 8-Core Processor 1.683,8 tok/sAMD Ryzen 9 9950X 16-Core Processor 1.670,5 tok/s★ AMD Ryzen 5 5600X 6-Core Processor 1.623,2 tok/s this runAMD Ryzen Threadripper PRO 9965WX 24-Cores 1.521,4 tok/sAMD Ryzen 9 8945HX with Radeon Graphics 930,6 tok/sAMD Ryzen Threadripper PRO 7955WX 16-Cores 751,7 tok/sAMD Ryzen Threadripper PRO 3955WX 16-Cores 303,1 tok/sAMD Ryzen Threadripper PRO 5975WX 32-Cores 258,3 tok/sAMD Ryzen 3 3100 4-Core Processor 238,8 tok/sIntel(R) Xeon(R) CPU E5-4657L v2 @ 2.40GHz 223,3 tok/sNVIDIA Grace 161,1 tok/sNVIDIA RTX Spark N1X (6144-core GPU, 20-core CPU) 107,0 tok/sIntel(R) Xeon(R) CPU E5-4620 v2 @ 2.60GHz 8,5 tok/s
MBby mainboard
★ ASUSTeK COMPUTER INC. ROG STRIX B550-A GAMING 1.683,8 tok/s this runASUSTeK COMPUTER INC. ProArt X870E-CREATOR WIFI 1.670,5 tok/sASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE 1.521,4 tok/sMeigao Innovation Technology (Shen Zhen) Co., Ltd DRFXL (MotherBoard Series) 930,6 tok/sASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI 303,1 tok/sASUSTeK COMPUTER INC. ROG STRIX X570-F GAMING 238,8 tok/sDell Inc. PowerEdge R820 223,3 tok/sASUSTeK COMPUTER INC. GX10 161,1 tok/sASUS H7607BA 107,0 tok/s
DRVby driver
unbekannt 1.683,8 tok/sNVIDIA 590.48.01 / CUDA 13.1 161,1 tok/sAMD 7.0.0-27-generic 91,4 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
12 W
⚡ TDP
362 W
☁️ External LLM (API)
Electricity0.30 EUR/kWh
Avg power (incl. idle)362 W estimated (TDP)GPU 350 + Board 12 W full load
Avg cost / hourEUR 0.11
Electricity / 1M tokensEUR 0.081
Token / kWh3.70M
Acquisition (system)EUR 1,518 full priceGPU EUR 850 · CPU EUR 121 · Board EUR 149 · RAM EUR 248 · PSU EUR 150
Electricity (2 years)–
TCO (2 years)EUR 3,421
Output tokens (2 years)23.44B
☁️ 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 (12 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
| Qwen3-30B-A3B-Thinking-2507NVIDIA GeForce RTX 3090 | Qwen3-30B-A3B-Thinking-2507NVIDIA GeForce RTX 5090 | Qwen3-30B-A3B-Thinking-2507NVIDIA RTX PRO 6000 Blackwell Workstation Edition | Qwen3-30B-A3B-Thinking-2507NVIDIA RTX PRO 6000 Blackwell 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 3090 · this run 372 tok/sNVIDIA GeForce RTX 5090 1,684 tok/sNVIDIA RTX PRO 6000 Blackwell Workstation Edition 1,670 tok/sNVIDIA RTX PRO 6000 Blackwell Workstation Edition 1,623 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.