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

Qwen2.5-32B-Instruct-AWQ

Performance benchmark · measured on 29.07.2026 07:04

Benchmark-IDrun-20260729-105333-bce47b
Timebench 3 - Kombi (Prefill + Generation)Dense32BRuntime: llama.cppQuantisierung: AWQ
Generation40,26tok/s
Prefill1.449,43tok/s
Time to First Token1.698,00ms
Total duration54,28s
Concurrency1parallel
Ranking in the field
294of 430 systems

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

This run is better than 32 % of all comparable systems.
Generation 40,3 tok/s
-68 % vs Ø 127,6
Prefill 1.449,4 tok/s
-57 % vs Ø 3.406,7
Time to First Token 1.698 ms
-80 % vs Ø 8.418
Distribution in the field0 – 405 tok/s
Ø 128 Median Dieser Lauf

Wie schlägt sich dieser Benchmark mit anderen Modellen?

gpt-oss-20bNVIDIA GeForce RTX 5090 · run-20260724-004608-16af9b
404,6 tok/s
NVIDIA-Nemotron-3-Nano-4BNVIDIA GeForce RTX 5090 · run-20260728-194135-d456e7
393,5 tok/s
NVIDIA-Nemotron-3-Nano-4BNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260727-181607-bffb11
393,5 tok/s
gpt-oss-20bNVIDIA GeForce RTX 5090 · run-20260724-004607-29bd16
388,9 tok/s
gpt-oss-20bNVIDIA GeForce RTX 5090 · run-20260729-032121-1c779f
388,2 tok/s
gemma-4-E2B-itNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260727-181606-6ea089
378,6 tok/s
gemma-4-E2B-itNVIDIA GeForce RTX 5090 · run-20260728-184455-e8b129
377,6 tok/s
NVIDIA-Nemotron-3-Nano-4B3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-194136-ac388e
362,3 tok/s
NVIDIA-Nemotron-3-Nano-4B3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-194136-5e5085
361,4 tok/s
NVIDIA-Nemotron-3-Nano-30B-A3BNVIDIA GeForce RTX 5090 · run-20260729-105335-653788
359,4 tok/s
Nemotron-3-Nano-Omni-30B-A3B-ReasoningNVIDIA GeForce RTX 5090 · run-20260729-105334-2c4905
359,3 tok/s
Nemotron-Cascade-2-30B-A3BNVIDIA GeForce RTX 5090 · run-20260729-105334-44d1e7
359,0 tok/s
Nemotron-3-Nano-Omni-30B-A3B-ReasoningNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260729-032119-4852bf
357,1 tok/s
NVIDIA-Nemotron-3-Nano-30B-A3BNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260729-032121-643b00
357,0 tok/s
Qwen2.5-32B-Instruct-AWQ this runNVIDIA GeForce RTX 3090 Ti · run-20260729-105333-bce47b
40,3 tok/s

How does this benchmark compare on other GPUs?

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

Configuration

benchmark-konfiguration — run-20260729-105333-bce47b
# LLM-Benchmark Konfiguration # Modell : Qwen2.5-32B-Instruct-AWQ # Engine : llama.cpp # Run-ID : run-20260729-105333-bce47b # GPU : NVIDIA GeForce RTX 3090 Ti # CPU : AMD Ryzen 9 8945HX with Radeon Graphics # RAM : 92 GB bench@llm-benchmark:~$ /root/llama.cpp/build/bin/llama-server \ -m /root/.cache/huggingface/hub/models--bartowski--Qwen2.5-32B-Instruct-GGUF/snapshots/2116cbb385b8ce3a4d28cf3bf1cd2039a55821a6/Qwen2.5-32B-Instruct-Q4_K_M.gguf \ --alias Qwen2.5-32B-Instruct-AWQ \ --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.Qwen2.5-32B-Instruct-AWQ
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./root/.cache/huggingface/hub/models--bartowski--Qwen2.5-32B-Instruct-GGUF/snapshots/2116cbb385b8ce3a4d28cf3bf1cd2039a55821a6/Qwen2.5-32B-Instruct-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
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

Qwen2.5-32B-Instruct-AWQ 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

6594943291650,001.8333.6675.500Prefill (tok/s)Generation (tok/s)NVIDIA RTX PRO 6000 Blackwell Workstation Edition - 510,3 tok/s Generation, 4.510 tok/s Prefill, TTFT 5.988 ms (6 Laufe)NVIDIA RTX PRO 6000 B...NVIDIA GeForce RTX 5070 Ti - 45,0 tok/s Generation, 619 tok/s Prefill, TTFT 95.102 ms (3 Laufe)NVIDIA GeForce RTX 50...AMD Radeon 8060S Graphics - 22,6 tok/s Generation, 517 tok/s Prefill, TTFT 24.115 ms (3 Laufe)AMD Radeon 8060S Grap...NVIDIA GeForce RTX 3090 Ti - 223,8 tok/s Generation, 2.049 tok/s Prefill, TTFT 20.166 ms (3 Laufe) | DIESER LAUF★ NVIDIA GeForce RTX 30...
NVIDIA RTX PRO 6000 Blackwell Workstation Edition 510,3 tok/s★ NVIDIA GeForce RTX 3090 Ti 223,8 tok/s this runNVIDIA GeForce RTX 5070 Ti 45,0 tok/sAMD Radeon 8060S Graphics 22,6 tok/s

CPUby processor

6594943291650,001.8333.6675.500Prefill (tok/s)Generation (tok/s)AMD Ryzen 9 9950X 16-Core Processor - 510,3 tok/s Generation, 4.510 tok/s Prefill, TTFT 5.988 ms (6 Laufe)AMD Ryzen 9 9950X 16-...AMD Ryzen Threadripper PRO 5975WX 32-Cores - 45,0 tok/s Generation, 619 tok/s Prefill, TTFT 95.102 ms (3 Laufe)AMD Ryzen Threadrippe...AMD RYZEN AI MAX+ 395 w/ Radeon 8060S - 22,6 tok/s Generation, 517 tok/s Prefill, TTFT 24.115 ms (3 Laufe)AMD RYZEN AI MAX+ 395...AMD Ryzen 9 8945HX with Radeon Graphics - 223,8 tok/s Generation, 2.049 tok/s Prefill, TTFT 20.166 ms (3 Laufe) | DIESER LAUF★ AMD Ryzen 9 8945HX wi...
AMD Ryzen 9 9950X 16-Core Processor 510,3 tok/s★ AMD Ryzen 9 8945HX with Radeon Graphics 223,8 tok/s this runAMD Ryzen Threadripper PRO 5975WX 32-Cores 45,0 tok/sAMD RYZEN AI MAX+ 395 w/ Radeon 8060S 22,6 tok/s

MBby mainboard

6594943291650,001.8333.6675.500Prefill (tok/s)Generation (tok/s)ASUSTeK COMPUTER INC. ProArt X870E-CREATOR WIFI - 510,3 tok/s Generation, 4.510 tok/s Prefill, TTFT 5.988 ms (6 Laufe)ASUSTeK COMPUTER INC....ASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI - 45,0 tok/s Generation, 619 tok/s Prefill, TTFT 95.102 ms (3 Laufe)ASUSTeK COMPUTER INC....Bosgame AXB35-02 (BeyondMax Series) - 22,6 tok/s Generation, 517 tok/s Prefill, TTFT 24.115 ms (3 Laufe)Bosgame AXB35-02 (Bey...Meigao Innovation Technology (Shen Zhen) Co., Ltd DRFXL (MotherBoard Series) - 223,8 tok/s Generation, 2.049 tok/s Prefill, TTFT 20.166 ms (3 Laufe) | DIESER LAUF★ Meigao Innovation Tec...
ASUSTeK COMPUTER INC. ProArt X870E-CREATOR WIFI 510,3 tok/s★ Meigao Innovation Technology (Shen Zhen) Co., Ltd DRFXL (MotherBoard Series) 223,8 tok/s this runASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI 45,0 tok/sBosgame AXB35-02 (BeyondMax Series) 22,6 tok/s

ENGby engine

5695304924544161.2192.5363.8545.171Prefill (tok/s)Generation (tok/s)vLLM - 510,3 tok/s Generation, 4.452 tok/s Prefill, TTFT 2.742 ms (3 Laufe)vLLMllama.cpp - 474,0 tok/s Generation, 1.939 tok/s Prefill, TTFT 37.154 ms (12 Laufe) | DIESER LAUF★ llama.cpp
vLLM 510,3 tok/s★ llama.cpp 474,0 tok/s this run

DRVby driver

5615365104854592.2952.3922.4902.588Prefill (tok/s)Generation (tok/s)unbekannt - 510,3 tok/s Generation, 2.441 tok/s Prefill, TTFT 30.272 ms (15 Laufe)unbekannt
unbekannt 510,3 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 50 W
⚡ TDP 477 W
☁️ External LLM (API)
Electricity0.30 EUR/kWh
Avg power (incl. idle)477 W estimated (TDP)GPU 450 + CPU 17 + Board 10 W full load
Avg cost / hourEUR 0.14
Electricity / 1M tokensEUR 0.99
Token / kWh303.69K
Acquisition (system)EUR 2,986 partial priceGPU EUR 999 · CPU EUR 549 · RAM EUR 1,288 · PSU EUR 150
Electricity (2 years)
TCO (2 years)EUR 5,494
Output tokens (2 years)2.54B
☁️ 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 (50 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

Qwen2.5-32B-Instruct-AWQNVIDIA GeForce RTX 3090 TiQwen2.5-32B-Instruct-AWQNVIDIA RTX PRO 6000 Blackwell Workstation EditionQwen2.5-32B-Instruct-AWQNVIDIA RTX PRO 6000 Blackwell Workstation EditionQwen2.5-32B-Instruct-AWQAMD Radeon 8060S Graphics
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.