Created bymario-alka.dePowered bygodcore.denoob2claw.detricoma.de
Contributed byMario AlkaQwen (Alibaba)

Qwen3-VL-30B-A3B-Instruct

Performance benchmark · measured on 18.08.2026 13:43

Benchmark-IDrun-20260818-141454-1e8f6c
Timebench 3 - Kombi (Prefill + Generation)MoE30BRuntime: llama.cppQuantisierung: Q8_0
Generation33,62tok/s
Prefill2.193,46tok/s
Time to First Token1.194,00ms
Total duration64,44s
Concurrency1parallel
Ranking in the field
604of 1161 systems

Performance benchmark · Primary metric: Generation-Speed (tok/s) · 1× concurrent

This run is better than 48 % of all comparable systems.
Generation 33,6 tok/s
-55 % vs Ø 74,3
Prefill 2.193,5 tok/s
-6 % vs Ø 2.324,7
Time to First Token 1.194 ms
-97 % vs Ø 35.657
Distribution in the field0 – 405 tok/s
Ø 74 Median Dieser Lauf

Wie schlägt sich dieser Benchmark mit anderen Modellen?

How does this benchmark compare on other GPUs?

Same model on different hardware · 1× concurrent · Generation (tok/s)

Hardware

GPU: 2x AMD Radeon PRO W7900 Dual Slot · 48 GB VRAM
CPU: AMD Ryzen Threadripper PRO 5975WX 32-Cores
RAM: 247 GB
Mainboard: ASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI

Setup

Runtime: llama.cpp
Quantization: Q8_0
Model: Qwen3-VL-30B-A3B-Instruct

Configuration

benchmark-konfiguration — run-20260818-141454-1e8f6c
# LLM-Benchmark Konfiguration # Modell : Qwen3-VL-30B-A3B-Instruct # Engine : llama.cpp # Run-ID : run-20260818-141454-1e8f6c # GPU : 2x AMD Radeon PRO W7900 Dual Slot # CPU : AMD Ryzen Threadripper PRO 5975WX 32-Cores # RAM : 247 GB bench@llm-benchmark:~$ /home/godcore/llama.cpp/build_hip/bin/llama-server \ -m /home/godcore/bench_scratch/Qwen_Qwen3-VL-30B-A3B-Instruct-Q8_0.gguf \ --alias Qwen3-VL-30B-A3B-Instruct \ -ngl 999 \ -fa on \ -sm layer \ --tensor-split 1,1 \ -c 49152 \ -np 12 \ --jinja \ --host 0.0.0.0 \ --port 8000
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.49152
Modellpfad?Pfad zur GGUF-Modelldatei, die geladen und ausgeliefert wird./home/godcore/bench_scratch/Qwen_Qwen3-VL-30B-A3B-Instruct-Q8_0.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
faon
Split-Mode?Verteilung ueber mehrere GPUs: none (nur eine GPU), layer (Layer aufteilen) oder row (Tensoren zeilenweise).layer
Tensor-Split?Verhaeltnis, in dem die Modell-Layer auf mehrere GPUs verteilt werden, z.B. 3,1.1,1
Kontext?Groesse des Kontextfensters in Token. 0 = der beim Training verwendete Kontext des Modells.49152
np12

All benchmarks of this model To leaderboard

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

2.1351.6011.0675340,006.96613.93220.898Prefill (tok/s)Generation (tok/s)NVIDIA GeForce RTX 5090 - 1.655,9 tok/s Generation, 9.318 tok/s Prefill, TTFT 3.329 ms (3 Laufe)NVIDIA GeForce RTX 50...NVIDIA RTX PRO 6000 Blackwell Workstation Edition - 1.639,0 tok/s Generation, 17.025 tok/s Prefill, TTFT 1.966 ms (6 Laufe)NVIDIA RTX PRO 6000 B...NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition - 1.505,0 tok/s Generation, 12.905 tok/s Prefill, TTFT 3.350 ms (9 Laufe)NVIDIA RTX PRO 6000 B...NVIDIA GeForce RTX 3090 Ti - 903,5 tok/s Generation, 5.934 tok/s Prefill, TTFT 5.817 ms (3 Laufe)NVIDIA GeForce RTX 30...NVIDIA GeForce RTX 5070 Ti - 154,0 tok/s Generation, 1.182 tok/s Prefill, TTFT 22.532 ms (6 Laufe)NVIDIA GeForce RTX 50...AMD Radeon AI PRO R9700 - 89,8 tok/s Generation, 3.530 tok/s Prefill, TTFT 44.809 ms (20 Laufe)AMD Radeon AI PRO R97...AMD Radeon PRO W7900 Dual Slot - 245,1 tok/s Generation, 4.091 tok/s Prefill, TTFT 3.139 ms (6 Laufe) | DIESER LAUF★ AMD Radeon PRO W7900 ...
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/s★ AMD Radeon PRO W7900 Dual Slot 245,1 tok/s this runNVIDIA GeForce RTX 5070 Ti 154,0 tok/sAMD Radeon AI PRO R9700 89,8 tok/s

CPUby processor

2.1351.6011.0675340,006.87913.75720.636Prefill (tok/s)Generation (tok/s)AMD Ryzen 7 5800X3D 8-Core Processor - 1.655,9 tok/s Generation, 9.318 tok/s Prefill, TTFT 3.329 ms (3 Laufe)AMD Ryzen 7 5800X3D 8...AMD Ryzen 9 9950X 16-Core Processor - 1.639,0 tok/s Generation, 17.025 tok/s Prefill, TTFT 1.966 ms (6 Laufe)AMD Ryzen 9 9950X 16-...AMD Ryzen Threadripper PRO 9965WX 24-Cores - 1.505,0 tok/s Generation, 12.905 tok/s Prefill, TTFT 3.350 ms (9 Laufe)AMD Ryzen Threadrippe...AMD Ryzen 9 8945HX with Radeon Graphics - 903,5 tok/s Generation, 5.934 tok/s Prefill, TTFT 5.817 ms (3 Laufe)AMD Ryzen 9 8945HX wi...AMD Ryzen Threadripper PRO 7955WX 16-Cores - 89,8 tok/s Generation, 3.530 tok/s Prefill, TTFT 44.809 ms (20 Laufe)AMD Ryzen Threadrippe...AMD Ryzen Threadripper PRO 5975WX 32-Cores - 245,1 tok/s Generation, 2.636 tok/s Prefill, TTFT 12.836 ms (12 Laufe) | DIESER LAUF★ AMD Ryzen Threadrippe...
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/s★ AMD Ryzen Threadripper PRO 5975WX 32-Cores 245,1 tok/s this runAMD Ryzen Threadripper PRO 7955WX 16-Cores 89,8 tok/s

MBby mainboard

2.1041.5781.0525260,006.87913.75720.636Prefill (tok/s)Generation (tok/s)ASUSTeK COMPUTER INC. ROG STRIX B550-A GAMING - 1.655,9 tok/s Generation, 9.318 tok/s Prefill, TTFT 3.329 ms (3 Laufe)ASUSTeK COMPUTER INC....ASUSTeK COMPUTER INC. ProArt X870E-CREATOR WIFI - 1.639,0 tok/s Generation, 17.025 tok/s Prefill, TTFT 1.966 ms (6 Laufe)ASUSTeK COMPUTER INC....ASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE - 1.505,0 tok/s Generation, 6.440 tok/s Prefill, TTFT 31.942 ms (29 Laufe)ASUSTeK COMPUTER INC....Meigao Innovation Technology (Shen Zhen) Co., Ltd DRFXL (MotherBoard Series) - 903,5 tok/s Generation, 5.934 tok/s Prefill, TTFT 5.817 ms (3 Laufe)Meigao Innovation Tec...ASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI - 245,1 tok/s Generation, 2.636 tok/s Prefill, TTFT 12.836 ms (12 Laufe) | DIESER LAUF★ ASUSTeK COMPUTER INC....
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/s★ ASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI 245,1 tok/s this run

ENGby engine

2.1351.6011.0675340,04.3596.6518.94211.233Prefill (tok/s)Generation (tok/s)vLLM - 1.028,5 tok/s Generation, 9.887 tok/s Prefill, TTFT 3.807 ms (10 Laufe)vLLMunbekannt - 89,8 tok/s Generation, 5.705 tok/s Prefill, TTFT 549 ms (7 Laufe)unbekanntllama.cpp - 1.655,9 tok/s Generation, 6.319 tok/s Prefill, TTFT 29.935 ms (36 Laufe) | DIESER LAUF★ llama.cpp
★ llama.cpp 1.655,9 tok/s this runvLLM 1.028,5 tok/sunbekannt 89,8 tok/s

DRVby driver

2.1351.6011.0675340,05.0295.9436.8577.770Prefill (tok/s)Generation (tok/s)unbekannt - 1.655,9 tok/s Generation, 7.095 tok/s Prefill, TTFT 24.255 ms (46 Laufe)unbekanntAMD 7.0.0-27-generic - 89,8 tok/s Generation, 5.705 tok/s Prefill, TTFT 549 ms (7 Laufe)AMD 7.0.0-27-generic
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 (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 10 W
⚡ TDP 600 W
☁️ External LLM (API)
Electricity0.30 EUR/kWh
Avg power (incl. idle)600 W estimated (TDP)GPU 590 + Board 10 W full load
Avg cost / hourEUR 0.18
Electricity / 1M tokensEUR 1.49
Token / kWh201.72K
Acquisition (system)EUR 2,156 missingRAM EUR 1,976 · PSU EUR 180
Electricity (2 years)
TCO (2 years)EUR 5,310
Output tokens (2 years)2.12B
☁️ 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-VL-30B-A3B-Instruct2x AMD Radeon PRO W7900 Dual SlotQwen3-VL-30B-A3B-InstructNVIDIA GeForce RTX 5090Qwen3-VL-30B-A3B-InstructNVIDIA RTX PRO 6000 Blackwell Workstation EditionQwen3-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)

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.