Contributed byMario AlkaGoogle gemma-4-12B-it
Performance benchmark · measured on 08.10.2026 08:42
Benchmark-ID
run-20261008-105921-638519Timebench 3 - Kombi (Prefill + Generation)Dense12BRuntime: llama.cppQuantisierung: Q4_K_M
ℹ
Public note: Dell PowerEdge R820 . 2x AMD Radeon RX 7900 XTX (48 GB VRAM gesamt, ROCm/HIP) . RAM: 32x 16 GB SK Hynix DDR3-1600 ECC RDIMM (HMT42GR7AFR4A-PB) = 512 GB @ 1600 MT/s
Generation51,06tok/s
Prefill1.138,08tok/s
Time to First Token1.778,50ms
Total duration43,67s
Concurrency1parallel
Ranking in the field
873of 1808 systems
Performance benchmark · Primary metric: Generation-Speed (tok/s) · 1× concurrent
This run is better than 52 % of all comparable systems.
Generation
51,1 tok/s
-34 % vs Ø 77,1
Prefill
1.138,1 tok/s
-60 % vs Ø 2.836,4
Time to First Token
1.779 ms
-93 % vs Ø 25.162
Distribution in the field0 – 405 tok/s
Wie schlägt sich dieser Benchmark mit anderen Modellen?
gpt-oss-20bNVIDIA GeForce RTX 5090 · run-20260724-004608-16af9b
404,6 tok/s
gpt-oss-20bNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260730-035052-9f766a
395,2 tok/s
Nemotron-3-Nano-4BNVIDIA GeForce RTX 5090 · run-20260728-194135-d456e7
393,5 tok/s
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
Nemotron-3-Nano-4B3× 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-194136-ac388e
362,3 tok/s
Nemotron-3-Nano-4B3× 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-194136-5e5085
361,4 tok/s
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
gemma-4-12B-it this run2× 2x AMD Radeon RX 7900 XTX · run-20261008-105921-638519
51,1 tok/s
How does this benchmark compare on other GPUs?
Same model on different hardware · 1× concurrent · Generation (tok/s)
NVIDIA GeForce RTX 5090run-20260729-060804-413d64
154,0 tok/s
NVIDIA RTX PRO 6000 Blackwell Workstation Editionrun-20260728-194137-861b96
151,5 tok/s
3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Editionrun-20260729-105332-034286
144,4 tok/s
3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Editionrun-20260729-105332-666d50
143,9 tok/s
3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Editionrun-20260729-105332-a880da
143,7 tok/s
NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Editionrun-20260923-143546-e524d3
126,1 tok/s
2× NVIDIA RTX A6000run-20260829-110114-9a8021
108,5 tok/s
2× NVIDIA RTX A6000run-20260827-073158-a8ccfd
99,2 tok/s
NVIDIA GeForce RTX 3090 Tirun-20260729-032127-d1f86f
93,1 tok/s
NVIDIA GeForce RTX 5070 Tirun-20260728-194136-e59252
92,8 tok/s
NVIDIA GeForce RTX 3090 Tirun-20260804-052132-ea8471
83,6 tok/s
2× NVIDIA RTX A6000run-20260829-113415-65be31
80,5 tok/s
2× NVIDIA RTX A6000run-20260827-083601-0afaeb
75,5 tok/s
NVIDIA GeForce RTX 5070 Tirun-20260727-032759-4364c2
75,0 tok/s
Hardware
GPU: 2x AMD Radeon RX 7900 XTX · 24 GB VRAM
CPU: 4x Intel(R) Xeon(R) CPU E5-4657L v2 @ 2.40GHz
RAM: 504 GB
Mainboard: Dell Inc. PowerEdge R820
Setup
Runtime: llama.cpp
Quantization: Q4_K_M
Model: gemma-4-12B-it
Configuration
# LLM-Benchmark Konfiguration
# Modell : gemma-4-12B-it
# Engine : llama.cpp
# Run-ID : run-20261008-105921-638519
# GPU : 2x AMD Radeon RX 7900 XTX
# CPU : 4x Intel(R) Xeon(R) CPU E5-4657L v2 @ 2.40GHz
# RAM : 504 GB
bench@llm-benchmark:~$ llama-server \
-m gemma-4-12B-it-IQ4_XS.gguf \
--alias gemma-4-12B-it \
-ngl 999 \
-sm layer \
-fa on \
-c 40960 \
-np 10 '(2x' RX 7900 XTX ROCm/HIP, 'gpu)'
| 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. | gemma-4-12B-it |
| Kontextlaenge?Maximale Anzahl Tokens (Eingabe + erzeugte Ausgabe zusammen), die das Modell pro Anfrage verarbeiten kann. | 40960 |
| Modellpfad?Pfad zur GGUF-Modelldatei, die geladen und ausgeliefert wird. | gemma-4-12B-it-IQ4_XS.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 |
| Split-Mode?Verteilung ueber mehrere GPUs: none (nur eine GPU), layer (Layer aufteilen) oder row (Tensoren zeilenweise). | layer |
| fa | on |
| Kontext?Groesse des Kontextfensters in Token. 0 = der beim Training verwendete Kontext des Modells. | 40960 |
| np | 10 |
| execution_type | local |
Anzeige
Model comparison
gemma-4-12B-it 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 RTX PRO 6000 Blackwell Workstation Edition 1.104,3 tok/sNVIDIA GeForce RTX 5090 1.058,9 tok/sNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition 1.041,8 tok/sNVIDIA RTX A6000 681,8 tok/sNVIDIA GeForce RTX 5070 Ti 677,5 tok/sNVIDIA GeForce RTX 3090 Ti 654,9 tok/s★ AMD Radeon RX 7900 XTX 212,9 tok/s this runAMD Radeon PRO W7800 48GB 202,0 tok/sAMD Radeon AI PRO R9700 188,8 tok/sNVIDIA Tesla V100 PCIe 32GB 186,2 tok/sNVIDIA GeForce RTX 2060 89,8 tok/sNVIDIA Tesla P100 PCIe 16GB 45,8 tok/s
CPUby processor
AMD Ryzen 9 9950X 16-Core Processor 1.104,3 tok/sAMD Ryzen 7 5800X3D 8-Core Processor 1.058,9 tok/sAMD Ryzen Threadripper PRO 9965WX 24-Cores 1.041,8 tok/sAMD Ryzen 9 7945HX with Radeon Graphics 861,4 tok/sAMD Ryzen Threadripper PRO 7955WX 16-Cores 681,8 tok/sAMD Ryzen Threadripper PRO 5975WX 32-Cores 677,5 tok/sAMD Ryzen 9 8945HX with Radeon Graphics 654,9 tok/sAMD Ryzen 5 5600X 6-Core Processor 372,4 tok/sAMD Ryzen Threadripper PRO 3955WX 16-Cores 212,9 tok/sAMD Ryzen 3 3100 4-Core Processor 188,8 tok/sAMD Ryzen 3 PRO 3200GE w/ Radeon Vega Graphics 186,2 tok/s★ Intel(R) Xeon(R) CPU E5-4657L v2 @ 2.40GHz 179,4 tok/s this runIntel(R) Core(TM) i5-7400 CPU @ 3.00GHz 89,8 tok/s
MBby mainboard
ASUSTeK COMPUTER INC. ProArt X870E-CREATOR WIFI 1.104,3 tok/sASUSTeK COMPUTER INC. ROG STRIX B550-A GAMING 1.058,9 tok/sASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE 1.041,8 tok/sShenzhen Meigao Electronic Equipment Co.,Ltd DRFXI (MotherBoard Series) 861,4 tok/sASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI 677,5 tok/sMeigao Innovation Technology (Shen Zhen) Co., Ltd DRFXL (MotherBoard Series) 654,9 tok/sASUSTeK COMPUTER INC. PRIME A520M-K 372,4 tok/sASUSTeK COMPUTER INC. ROG STRIX X570-F GAMING 188,8 tok/s★ Dell Inc. PowerEdge R820 179,4 tok/s this runASRock H110 Pro BTC+ 89,8 tok/sShenzhen Meigao Electronic Equipment Co.,Ltd F1FXM (DeskMini Series) 45,8 tok/s
DRVby driver
unbekannt 1.104,3 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
35 W
⚡ TDP
749 W
☁️ External LLM (API)
Electricity0.30 EUR/kWh
Avg power (incl. idle)749 W estimated (TDP)GPU 710 + CPU 29 + Board 10 W full load
Avg cost / hourEUR 0.22
Electricity / 1M tokensEUR 1.22
Token / kWh245.50K
Acquisition (system)EUR 4,353 partial priceGPU EUR 2,098 · CPU EUR 59 · RAM EUR 2,016 · PSU EUR 180
Electricity (2 years)–
TCO (2 years)EUR 8,288
Output tokens (2 years)3.22B
☁️ 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 (35 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
| gemma-4-12B-it2x AMD Radeon RX 7900 XTX | gemma-4-12B-itNVIDIA GeForce RTX 5090 | gemma-4-12B-itNVIDIA RTX PRO 6000 Blackwell Workstation Edition | gemma-4-12B-it3x 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) | – | – | – | – |
2× AMD Radeon RX 7900 XTX · this run 51 tok/sNVIDIA GeForce RTX 5090 154 tok/sNVIDIA RTX PRO 6000 Blackwell Workstation Edition 152 tok/s3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition 144 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.