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

Qwen3-30B-A3B-Thinking-2507

Performance benchmark · measured on 27.07.2026 06:38

Benchmark-IDrun-20260727-082116-78fc9e
Timebench 3 - Kombi (Prefill + Generation)MoE30BRuntime: vLLMQuantisierung: AWQ
For context: Teile des Modells wurden per --cpu-offload-gb=80 GB in den System-RAM ausgelagert – RAM wird als erweiterter VRAM genutzt, was den Durchsatz senken kann.
Generation29,62tok/s
Prefill1.223,17tok/s
Time to First Token11.694,50ms
Total duration369,93s
Concurrency5parallel
Ranking in the field
1054of 1234 systems

Performance benchmark · Primary metric: Generation-Speed (tok/s) · 5× concurrent

This run is better than 15 % of all comparable systems.
Generation 29,6 tok/s
-89 % vs Ø 260,2
Prefill 1.223,2 tok/s
-77 % vs Ø 5.330,6
Time to First Token 11.695 ms
-61 % vs Ø 29.782
Distribution in the field0 – 1.349 tok/s
Ø 260 Median Dieser Lauf

Wie schlägt sich dieser Benchmark mit anderen Modellen?

gemma-4-E2B-itNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260727-181606-0adf6d
1.349,3 tok/s
gemma-4-E2B-itNVIDIA GeForce RTX 5090 · run-20260728-184455-5b4937
1.301,9 tok/s
gemma-4-E2B-it3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-184456-4111a9
1.195,7 tok/s
Nemotron-3-Nano-4BNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260727-181607-30d864
1.188,7 tok/s
Nemotron-3-Nano-4BNVIDIA GeForce RTX 5090 · run-20260728-194135-5432cd
1.164,8 tok/s
gpt-oss-20bNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260728-140954-9e81f9
1.136,3 tok/s
gpt-oss-20bNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260730-035052-09627f
1.135,0 tok/s
gpt-oss-20bNVIDIA GeForce RTX 5090 · run-20260729-032121-058a31
1.113,5 tok/s
Nemotron-3-Nano-4B3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-194136-836a91
1.090,6 tok/s
Nemotron-3-Nano-4B3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260727-181606-8b381e
1.077,4 tok/s
Nemotron-3-Nano-4B3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-194136-b8a818
1.071,0 tok/s
Nemotron-3-Nano-4B3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-194136-038e92
1.051,1 tok/s
gpt-oss-20b3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260729-032121-ffb18a
1.015,5 tok/s
Nemotron-3-Nano-30B-A3BNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260729-032121-305ad6
1.007,8 tok/s
Qwen3-30B-A3B-Thinking-2507 this runNVIDIA GeForce RTX 5070 Ti · run-20260727-082116-78fc9e
29,6 tok/s

How does this benchmark compare on other GPUs?

Same model on different hardware · 5× concurrent · Generation (tok/s)

Hardware

GPU: NVIDIA GeForce RTX 5070 Ti · 16 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: vLLM
Quantization: AWQ
Model: Qwen3-30B-A3B-Thinking-2507

Configuration

benchmark-konfiguration — run-20260727-082116-78fc9e
# LLM-Benchmark Konfiguration # Modell : Qwen3-30B-A3B-Thinking-2507 # Engine : vLLM # Run-ID : run-20260727-082116-78fc9e # GPU : NVIDIA GeForce RTX 5070 Ti # CPU : AMD Ryzen Threadripper PRO 5975WX 32-Cores # RAM : 247 GB bench@llm-benchmark:~$ /home/godcore/vllm-venv/bin/python3 /home/godcore/vllm-venv/bin/vllm serve cyankiwi/Qwen3-30B-A3B-Thinking-2507-AWQ-4bit \ --served-model-name Qwen3-30B-A3B-Thinking-2507 \ --tensor-parallel-size 1 \ --cpu-offload-gb 80 \ --dtype auto \ --max-model-len 4096 \ --enforce-eager \ --gpu-memory-utilization 0.60 \ --trust-remote-code \ --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.vllm
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.4096
Alias?Anzeigename des Modells nach aussen (served model name), unabhaengig vom Dateinamen.Qwen3-30B-A3B-Thinking-2507
Tensor-Parallel?Anzahl GPUs, auf die JEDER einzelne Modell-Layer aufgeteilt wird (Tensor-Parallelitaet). Mehr GPUs = mehr VRAM und meist mehr Speed, aber die Anzahl der Attention-Heads muss durch diesen Wert teilbar sein (z.B. 32 Heads -> nur 1, 2, 4, 8 ... moeglich, NICHT 3).1
cpu-offload-gb?Menge an Modellgewichten in GiB, die in den CPU-RAM ausgelagert wird. Ermoeglicht groessere Modelle als der GPU-Speicher fasst, kostet aber Geschwindigkeit.80
Dtype?Zahlenformat der Modellgewichte bei der Berechnung (z.B. auto, float16, bfloat16). 'auto' waehlt automatisch das vom Modell empfohlene Format.auto
Kontext?Groesse des Kontextfensters in Token. 0 = der beim Training verwendete Kontext des Modells.4096
GPU-Speicher?Anteil des GPU-Speichers (0 bis 1), den vLLM belegen darf. 0.92 = 92 %. Hoeher = mehr Platz fuer den KV-Cache (mehr/laengere parallele Anfragen), aber groesseres Risiko fuer 'Out of Memory'.0.60

All benchmarks of this model To leaderboard

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

2.1871.6401.0945470,007.58515.17022.756Prefill (tok/s)Generation (tok/s)NVIDIA GeForce RTX 5090 - 1.683,8 tok/s Generation, 14.349 tok/s Prefill, TTFT 2.214 ms (6 Laufe)NVIDIA GeForce RTX 50...NVIDIA RTX PRO 6000 Blackwell Workstation Edition - 1.670,5 tok/s Generation, 18.362 tok/s Prefill, TTFT 1.942 ms (6 Laufe)NVIDIA RTX PRO 6000 B...NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition - 1.521,4 tok/s Generation, 10.574 tok/s Prefill, TTFT 3.521 ms (9 Laufe)NVIDIA RTX PRO 6000 B...NVIDIA GeForce RTX 3090 Ti - 930,6 tok/s Generation, 7.726 tok/s Prefill, TTFT 5.880 ms (6 Laufe)NVIDIA GeForce RTX 30...NVIDIA RTX A6000 - 751,7 tok/s Generation, 14.399 tok/s Prefill, TTFT 1.684 ms (36 Laufe)NVIDIA RTX A6000AMD Radeon PRO W7800 48GB - 258,3 tok/s Generation, 4.982 tok/s Prefill, TTFT 2.607 ms (3 Laufe)AMD Radeon PRO W7800 ...AMD Radeon PRO W7900 Dual Slot - 254,6 tok/s Generation, 4.711 tok/s Prefill, TTFT 2.915 ms (6 Laufe)AMD Radeon PRO W7900 ...AMD Radeon AI PRO R9700 - 91,4 tok/s Generation, 2.676 tok/s Prefill, TTFT 39.171 ms (25 Laufe)AMD Radeon AI PRO R97...NVIDIA GeForce RTX 2060 - 8,5 tok/s Generation, 74 tok/s Prefill, TTFT 134.321 ms (3 Laufe)NVIDIA GeForce RTX 20...NVIDIA GeForce RTX 5070 Ti - 155,8 tok/s Generation, 1.078 tok/s Prefill, TTFT 23.667 ms (6 Laufe) | DIESER LAUF★ NVIDIA GeForce RTX 50...
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/sAMD Radeon PRO W7800 48GB 258,3 tok/sAMD Radeon PRO W7900 Dual Slot 254,6 tok/s★ NVIDIA GeForce RTX 5070 Ti 155,8 tok/s this runAMD Radeon AI PRO R9700 91,4 tok/sNVIDIA GeForce RTX 2060 8,5 tok/s

CPUby processor

2.1871.6401.0945470,007.58515.17022.756Prefill (tok/s)Generation (tok/s)AMD Ryzen 7 5800X3D 8-Core Processor - 1.683,8 tok/s Generation, 14.349 tok/s Prefill, TTFT 2.214 ms (6 Laufe)AMD Ryzen 7 5800X3D 8...AMD Ryzen 9 9950X 16-Core Processor - 1.670,5 tok/s Generation, 18.362 tok/s Prefill, TTFT 1.942 ms (6 Laufe)AMD Ryzen 9 9950X 16-...AMD Ryzen Threadripper PRO 9965WX 24-Cores - 1.521,4 tok/s Generation, 10.574 tok/s Prefill, TTFT 3.521 ms (9 Laufe)AMD Ryzen Threadrippe...AMD Ryzen 9 8945HX with Radeon Graphics - 930,6 tok/s Generation, 7.726 tok/s Prefill, TTFT 5.880 ms (6 Laufe)AMD Ryzen 9 8945HX wi...AMD Ryzen Threadripper PRO 7955WX 16-Cores - 751,7 tok/s Generation, 9.595 tok/s Prefill, TTFT 17.048 ms (61 Laufe)AMD Ryzen Threadrippe...Intel(R) Xeon(R) CPU E5-4620 v2 @ 2.60GHz - 8,5 tok/s Generation, 74 tok/s Prefill, TTFT 134.321 ms (3 Laufe)Intel(R) Xeon(R) CPU ...AMD Ryzen Threadripper PRO 5975WX 32-Cores - 258,3 tok/s Generation, 3.312 tok/s Prefill, TTFT 11.154 ms (15 Laufe) | DIESER LAUF★ AMD Ryzen Threadrippe...
AMD Ryzen 7 5800X3D 8-Core Processor 1.683,8 tok/sAMD Ryzen 9 9950X 16-Core Processor 1.670,5 tok/sAMD 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/s★ AMD Ryzen Threadripper PRO 5975WX 32-Cores 258,3 tok/s this runIntel(R) Xeon(R) CPU E5-4620 v2 @ 2.60GHz 8,5 tok/s

MBby mainboard

2.1871.6401.0945470,007.58515.17022.756Prefill (tok/s)Generation (tok/s)ASUSTeK COMPUTER INC. ROG STRIX B550-A GAMING - 1.683,8 tok/s Generation, 14.349 tok/s Prefill, TTFT 2.214 ms (6 Laufe)ASUSTeK COMPUTER INC....ASUSTeK COMPUTER INC. ProArt X870E-CREATOR WIFI - 1.670,5 tok/s Generation, 18.362 tok/s Prefill, TTFT 1.942 ms (6 Laufe)ASUSTeK COMPUTER INC....ASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE - 1.521,4 tok/s Generation, 9.721 tok/s Prefill, TTFT 15.308 ms (70 Laufe)ASUSTeK COMPUTER INC....Meigao Innovation Technology (Shen Zhen) Co., Ltd DRFXL (MotherBoard Series) - 930,6 tok/s Generation, 7.726 tok/s Prefill, TTFT 5.880 ms (6 Laufe)Meigao Innovation Tec...Dell Inc. PowerEdge R820 - 8,5 tok/s Generation, 74 tok/s Prefill, TTFT 134.321 ms (3 Laufe)Dell Inc. PowerEdge R...ASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI - 258,3 tok/s Generation, 3.312 tok/s Prefill, TTFT 11.154 ms (15 Laufe) | DIESER LAUF★ ASUSTeK COMPUTER INC....
ASUSTeK COMPUTER INC. ROG STRIX B550-A GAMING 1.683,8 tok/sASUSTeK 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/s★ ASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI 258,3 tok/s this runDell Inc. PowerEdge R820 8,5 tok/s

ENGby engine

2.1711.6281.0855430,01.0626.90212.74218.583Prefill (tok/s)Generation (tok/s)llama.cpp - 1.683,8 tok/s Generation, 6.274 tok/s Prefill, TTFT 25.535 ms (63 Laufe)llama.cppunbekannt - 91,4 tok/s Generation, 4.068 tok/s Prefill, TTFT 1.364 ms (8 Laufe)unbekanntvLLM - 906,0 tok/s Generation, 15.577 tok/s Prefill, TTFT 2.356 ms (35 Laufe) | DIESER LAUF★ vLLM
llama.cpp 1.683,8 tok/s★ vLLM 906,0 tok/s this rununbekannt 91,4 tok/s

DRVby driver

2.1711.6281.0855430,02.4975.3878.27711.167Prefill (tok/s)Generation (tok/s)unbekannt - 1.683,8 tok/s Generation, 9.596 tok/s Prefill, TTFT 17.257 ms (98 Laufe)unbekanntAMD 7.0.0-27-generic - 91,4 tok/s Generation, 4.068 tok/s Prefill, TTFT 1.364 ms (8 Laufe)AMD 7.0.0-27-generic
unbekannt 1.683,8 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 (5× 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 310 W
☁️ External LLM (API)
Electricity0.30 EUR/kWh
Avg power (incl. idle)310 W estimated (TDP)GPU 300 + Board 10 W full load
Avg cost / hourEUR 0.093
Electricity / 1M tokensEUR 0.87
Token / kWh343.97K
Acquisition (system)EUR 5,000 partial priceGPU EUR 994 · CPU EUR 1,880 · RAM EUR 1,976 · PSU EUR 150
Electricity (2 years)
TCO (2 years)EUR 6,629
Output tokens (2 years)1.87B
☁️ 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

Qwen3-30B-A3B-Thinking-2507NVIDIA GeForce RTX 5070 TiQwen3-30B-A3B-Thinking-2507NVIDIA GeForce RTX 5090Qwen3-30B-A3B-Thinking-2507NVIDIA RTX PRO 6000 Blackwell Workstation EditionQwen3-30B-A3B-Thinking-25073x 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.