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

Qwen3.5-122B-A10B-Q8

Performance benchmark · measured on 01.08.2026 11:14

Benchmark-IDrun-20260803-043843-5beddc
Timebench 3 - Kombi (Prefill + Generation)MoE122BRuntime: llama.cppQuantisierung: Q8_0
Generation549,60tok/s
Prefill3.637,44tok/s
Time to First Token19.950,00ms
Total duration119,87s
Concurrency10parallel
Ranking in the field
116of 204 systems

Performance benchmark · Primary metric: Generation-Speed (tok/s) · 10× concurrent

This run is better than 43 % of all comparable systems.
Generation 549,6 tok/s
-35 % vs Ø 849,0
Prefill 3.637,4 tok/s
-47 % vs Ø 6.900,9
Time to First Token 19.950 ms
-35 % vs Ø 30.862
Distribution in the field3 – 2.144 tok/s
Ø 849 Median Dieser Lauf

Wie schlägt sich dieser Benchmark mit anderen Modellen?

gemma-4-E2B-it3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-184456-d87c73
2.143,6 tok/s
NVIDIA-Nemotron-3-Nano-4B3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-194136-0aed86
1.937,7 tok/s
NVIDIA-Nemotron-3-Nano-4B3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-194136-8db0fa
1.911,6 tok/s
NVIDIA-Nemotron-3-Nano-4B3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260727-181606-e0eafd
1.897,0 tok/s
gpt-oss-20b3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-034052-26063b
1.835,7 tok/s
NVIDIA-Nemotron-3-Nano-4B3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-194136-a77046
1.830,3 tok/s
gpt-oss-20b3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260729-032121-c6e7e0
1.819,5 tok/s
gpt-oss-20b3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260729-032121-2b99aa
1.802,0 tok/s
Ornith-1.0-35B3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260731-071556-e9a354
1.799,4 tok/s
gpt-oss-20b3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260729-032120-1cc62d
1.742,1 tok/s
Nemotron-Cascade-2-30B-A3B3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260729-105340-3fa079
1.655,2 tok/s
NVIDIA-Nemotron-3-Nano-30B-A3B3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260729-105341-be087a
1.654,6 tok/s
Nemotron-3-Nano-Omni-30B-A3B-Reasoning3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260729-105340-2fab14
1.651,0 tok/s
Nemotron-Cascade-2-30B-A3B3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260729-105340-b8a790
1.650,7 tok/s
Qwen3.5-122B-A10B-Q8 this run3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260803-043843-5beddc
549,6 tok/s

Hardware

GPU: 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 96 GB VRAM
CPU: AMD Ryzen Threadripper PRO 9965WX 24-Cores
RAM: 125 GB
Mainboard: ASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE

Setup

Runtime: llama.cpp
Quantization: Q8_0
Model: Qwen3.5-122B-A10B-Q8

Configuration

benchmark-konfiguration — run-20260803-043843-5beddc
# LLM-Benchmark Konfiguration # Modell : Qwen3.5-122B-A10B-Q8 # Engine : llama.cpp # Run-ID : run-20260803-043843-5beddc # GPU : 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition # CPU : AMD Ryzen Threadripper PRO 9965WX 24-Cores # RAM : 125 GB bench@llm-benchmark:~$ /home/godcore/llama.cpp/build/bin/llama-server \ -m /home/godcore/.cache/huggingface/hub/models--unsloth--Qwen3.5-122B-A10B-GGUF/snapshots/51eab4d59d53f573fb9206cb3ce613f1d0aa392b/Q8_0/Qwen3.5-122B-A10B-Q8_0-00001-of-00004.gguf \ --alias Qwen3.5-122B-A10B-Q8 \ --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.Qwen3.5-122B-A10B-Q8
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./home/godcore/.cache/huggingface/hub/models--unsloth--Qwen3.5-122B-A10B-GGUF/snapshots/51eab4d59d53f573fb9206cb3ce613f1d0aa392b/Q8_0/Qwen3.5-122B-A10B-Q8_0-00001-of-00004.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
np4

All benchmarks of this model To leaderboard

Anzeige
Model comparison

Qwen3.5-122B-A10B-Q8 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

6175895615335051.8271.9051.9832.061Prefill (tok/s)Generation (tok/s)NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition - 561,1 tok/s Generation, 1.944 tok/s Prefill, TTFT 40.330 ms (9 Laufe) | DIESER LAUF★ NVIDIA RTX PRO 6000 B...
★ NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition 561,1 tok/s this run

CPUby processor

6175895615335051.8271.9051.9832.061Prefill (tok/s)Generation (tok/s)AMD Ryzen Threadripper PRO 9965WX 24-Cores - 561,1 tok/s Generation, 1.944 tok/s Prefill, TTFT 40.330 ms (9 Laufe) | DIESER LAUF★ AMD Ryzen Threadrippe...
★ AMD Ryzen Threadripper PRO 9965WX 24-Cores 561,1 tok/s this run

MBby mainboard

6175895615335051.8271.9051.9832.061Prefill (tok/s)Generation (tok/s)ASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE - 561,1 tok/s Generation, 1.944 tok/s Prefill, TTFT 40.330 ms (9 Laufe) | DIESER LAUF★ ASUSTeK COMPUTER INC....
★ ASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE 561,1 tok/s this run

ENGby engine

6175895615335051.8271.9051.9832.061Prefill (tok/s)Generation (tok/s)llama.cpp - 561,1 tok/s Generation, 1.944 tok/s Prefill, TTFT 40.330 ms (9 Laufe) | DIESER LAUF★ llama.cpp
★ llama.cpp 561,1 tok/s this run

DRVby driver

6175895615335051.8271.9051.9832.061Prefill (tok/s)Generation (tok/s)unbekannt - 561,1 tok/s Generation, 1.944 tok/s Prefill, TTFT 40.330 ms (9 Laufe)unbekannt
unbekannt 561,1 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 165 W
⚡ TDP 983 W
☁️ External LLM (API)
Electricity0.30 EUR/kWh
Avg power (incl. idle)983 W estimated (TDP)GPU 900 + CPU 57 + Board 25 W full load
Avg cost / hourEUR 0.29
Electricity / 1M tokensEUR 0.15
Token / kWh2.01M
Acquisition (system)EUR 45,748 full priceGPU EUR 39,000 · CPU EUR 3,499 · Board EUR 1,299 · RAM EUR 1,750 · PSU EUR 200
Electricity (2 years)
TCO (2 years)EUR 50,912
Output tokens (2 years)34.66B
☁️ 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 (165 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

Qwen3.5-122B-A10B-Q83x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionQwen3.5-122B-A10B-Q83x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionQwen3.5-122B-A10B-Q83x 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.