ℹ

For context: Diese Plattform nutzt Unified Memory – die "VRAM" ist gemeinsamer System-RAM (APU/Superchip); das Modell teilt sich den Speicher mit dem System.

Generation826,24tok/s

Prefill3.229,13tok/s

Time to First Token15.222,50ms

Total duration85,25s

Concurrency10parallel

Ranking in the field

This GPU All GPUs

173of 433 systems

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

This run is better than 60 % of all comparable systems.

Generation 826,2 tok/s

+16 % vs Ø 709,5

Prefill 3.229,1 tok/s

-52 % vs Ø 6.790,5

Time to First Token 15.223 ms

-62 % vs Ø 40.103

Distribution in the field0 – 2.491 tok/s

Ø 710 Median Dieser Lauf

Wie schlägt sich dieser Benchmark mit anderen Modellen?

gemma-4-E2B-it same modelNVIDIA GeForce RTX 5090 · run-20260728-184455-e28775

2.491,2 tok/s

gemma-4-E2B-it same modelNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260727-181606-b37ed9

2.414,4 tok/s

NVIDIA-Nemotron-3-Nano-4BNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260727-181607-4cca42

2.182,7 tok/s

gemma-4-E2B-it same model3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-184456-d87c73

2.143,6 tok/s

gpt-oss-20bNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260728-140954-8007c4

1.993,5 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-20bNVIDIA GeForce RTX 5090 · run-20260729-032121-7c961d

1.890,7 tok/s

NVIDIA-Nemotron-3-Nano-4BNVIDIA GeForce RTX 5090 · run-20260728-194135-2ed9dd

1.878,7 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

gemma-4-E2B-it this run2× AMD Radeon Graphics · run-20260729-224611-0989d2

826,2 tok/s

How does this benchmark compare on other GPUs?

Same model on different hardware · 10× concurrent · Generation (tok/s)

NVIDIA GeForce RTX 5090run-20260728-184455-e28775

2.491,2 tok/s

NVIDIA RTX PRO 6000 Blackwell Workstation Editionrun-20260727-181606-b37ed9

2.414,4 tok/s

3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Editionrun-20260728-184456-d87c73

2.143,6 tok/s

NVIDIA GeForce RTX 5070 Tirun-20260728-140953-875f2e

1.744,0 tok/s

NVIDIA GeForce RTX 5070 Tirun-20260727-090152-d5613d

1.743,1 tok/s

NVIDIA GeForce RTX 5070 Tirun-20260727-032758-bb6a92

1.608,5 tok/s

NVIDIA GeForce RTX 3090 Tirun-20260727-142111-067e7e

1.492,8 tok/s

NVIDIA GB10run-20260729-224611-cc3776

903,3 tok/s

NVIDIA GB10run-20260729-224610-32ded2

860,9 tok/s

2× AMD Radeon Graphics this runrun-20260729-224611-0989d2

826,2 tok/s

3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Editionrun-20260728-184456-36b887

274,8 tok/s

3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Editionrun-20260728-184456-294d9c

273,6 tok/s

3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Editionrun-20260727-155936-b597f5

266,1 tok/s

Hardware

GPU: 2x AMD Radeon Graphics
CPU: AMD Ryzen 9 7945HX with Radeon Graphics
RAM: 29 GB
Mainboard: Shenzhen Meigao Electronic Equipment Co.,Ltd DRFXI (MotherBoard Series)

Setup

Runtime: llama.cpp
Quantization: Q4_K_M
Model: gemma-4-E2B-it

Configuration

benchmark-konfiguration — run-20260729-224611-0989d2

# LLM-Benchmark Konfiguration # Modell : gemma-4-E2B-it # Engine : llama.cpp # Run-ID : run-20260729-224611-0989d2 # GPU : 2x AMD Radeon Graphics # CPU : AMD Ryzen 9 7945HX with Radeon Graphics # RAM : 29 GB bench@llm-benchmark:~$ /root/llama.cpp/build/bin/llama-server \ -m /root/.cache/huggingface/hub/models--unsloth--gemma-4-E2B-it-GGUF/snapshots/0314792d7f1f7e229411f620751375812bb9faf2/gemma-4-E2B-it-Q4_K_M.gguf \ --alias gemma-4-E2B-it \ --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.	gemma-4-E2B-it
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--unsloth--gemma-4-E2B-it-GGUF/snapshots/0314792d7f1f7e229411f620751375812bb9faf2/gemma-4-E2B-it-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
np	4

All benchmarks of this model To leaderboard

Model comparison

gemma-4-E2B-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 GeForce RTX 5090 2.491,2 tok/sNVIDIA RTX PRO 6000 Blackwell Workstation Edition 2.414,4 tok/sNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition 2.143,6 tok/sNVIDIA GeForce RTX 5070 Ti 1.744,0 tok/sNVIDIA GeForce RTX 3090 Ti 1.492,8 tok/sNVIDIA GB10 (DGX Spark) 903,3 tok/s★ AMD Radeon Graphics 826,2 tok/s this run

CPUby processor

AMD Ryzen 7 5800X3D 8-Core Processor 2.491,2 tok/sAMD Ryzen 9 9950X 16-Core Processor 2.414,4 tok/sAMD Ryzen Threadripper PRO 9965WX 24-Cores 2.143,6 tok/sAMD Ryzen Threadripper PRO 5975WX 32-Cores 1.744,0 tok/sAMD Ryzen 9 8945HX with Radeon Graphics 1.492,8 tok/sNVIDIA Grace 903,3 tok/s★ AMD Ryzen 9 7945HX with Radeon Graphics 826,2 tok/s this run

MBby mainboard

ASUSTeK COMPUTER INC. ROG STRIX B550-A GAMING 2.491,2 tok/sASUSTeK COMPUTER INC. ProArt X870E-CREATOR WIFI 2.414,4 tok/sASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE 2.143,6 tok/sASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI 1.744,0 tok/sMeigao Innovation Technology (Shen Zhen) Co., Ltd DRFXL (MotherBoard Series) 1.492,8 tok/sASUSTeK COMPUTER INC. GX10 903,3 tok/s★ Shenzhen Meigao Electronic Equipment Co.,Ltd DRFXI (MotherBoard Series) 826,2 tok/s this run

ENGby engine

★ llama.cpp 2.491,2 tok/s this runvLLM 1.608,5 tok/s

DRVby driver

unbekannt 2.491,2 tok/sNVIDIA 590.48.01 / CUDA 13.1 903,3 tok/s

💰 Economics

Economics of this run

Operating cost, TCO and comparison with the next-best runs of the same model at identical concurrency (10× concurrent). Methodology →

⚙️ ConfigurationAll metrics and charts below follow these settings – based on a 24-month runtime.Save to URL Reset

⚡ Electricity price EUR/kWh

⚙️ System utilization 100 %

🔌 Idle 0 W

⚡ TDP 10 W

☁️ External LLM (API) show

Input €/1M Output €/1M Input per output token

Electricity0.30 EUR/kWh

Avg power (incl. idle)0 W missing

Avg cost / hour–

Electricity / 1M tokens–

Token / kWh–

Acquisition (system)– missing

Electricity (2 years)–

TCO (2 years)–

Output tokens (2 years)52.11B

☁️ External LLM (API) – comparison

External LLM cost (2 years)–

Savings vs. external (2 years)–

No power draw measured – values estimated from GPU TDP + CPU (idle + 15 %) + board.

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 (0 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-E2B-it2x AMD Radeon Graphics	gemma-4-E2B-itNVIDIA GeForce RTX 5090	gemma-4-E2B-itNVIDIA RTX PRO 6000 Blackwell Workstation Edition	gemma-4-E2B-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 Graphics · this run 826 tok/s NVIDIA GeForce RTX 5090 2,491 tok/s NVIDIA RTX PRO 6000 Blackwell Workstation Edition 2,414 tok/s 3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition 2,144 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.