Contributed byMarcel SommerGoogle gemma-4-E2B-it
Performance benchmark · measured on 31.07.2026 11:59
Benchmark-ID
run-20260804-052131-b12338Timebench 3 - Kombi (Prefill + Generation)Dense5BRuntime: vLLMQuantisierung: BF16
Generation46,01tok/s
Prefill11.757,32tok/s
Time to First Token169,00ms
Total duration44,87s
Concurrency1parallel
Ranking in the field
49of 112 systems
Performance benchmark · Primary metric: Generation-Speed (tok/s) · 1× concurrent
This run is better than 57 % of all comparable systems.
Generation
46,0 tok/s
-35 % vs Ø 71,1
Prefill
11.757,3 tok/s
+445 % vs Ø 2.158,0
Time to First Token
169 ms
-100 % vs Ø 127.988
Distribution in the field1 – 246 tok/s
Wie schlägt sich dieser Benchmark mit anderen Modellen?
gpt-oss-20bNVIDIA GeForce RTX 3090 Ti · run-20260724-004605-36d393
246,1 tok/s
gpt-oss-20bNVIDIA GeForce RTX 3090 Ti · run-20260724-004605-845608
245,6 tok/s
gpt-oss-20bNVIDIA GeForce RTX 3090 Ti · run-20260728-034052-31fe3e
238,9 tok/s
Nemotron-3-Nano-4BNVIDIA GeForce RTX 3090 Ti · run-20260727-142111-7e0547
236,9 tok/s
gemma-4-E2B-it same modelNVIDIA GeForce RTX 3090 Ti · run-20260727-142111-fb08fa
235,8 tok/s
Nemotron-3-Nano-30B-A3BNVIDIA GeForce RTX 3090 Ti · run-20260729-060802-e36a50
222,0 tok/s
Nemotron-3-Nano-Omni-30B-A3B-ReasoningNVIDIA GeForce RTX 3090 Ti · run-20260729-035835-89abde
221,8 tok/s
Nemotron-Cascade-2-30B-A3BNVIDIA GeForce RTX 3090 Ti · run-20260729-060801-1e5cba
221,8 tok/s
Qwen3-30B-A3B-Thinking-2507NVIDIA GeForce RTX 3090 Ti · run-20260729-060803-28852f
198,8 tok/s
Qwen3-Coder-30B-A3B-InstructNVIDIA GeForce RTX 3090 Ti · run-20260729-060804-5f5c60
198,4 tok/s
Laguna-XS-2.1NVIDIA GeForce RTX 3090 Ti · run-20260728-140956-c96921
198,3 tok/s
Laguna-XS-2.1NVIDIA GeForce RTX 3090 Ti · run-20260728-140956-70297b
198,3 tok/s
Laguna-XS-2.1NVIDIA GeForce RTX 3090 Ti · run-20260728-140956-35f5ee
198,3 tok/s
Qwen3-Omni-30B-A3B-ThinkingNVIDIA GeForce RTX 3090 Ti · run-20260729-060805-5e22b7
198,1 tok/s
gemma-4-E2B-it this runNVIDIA GeForce RTX 3090 Ti · run-20260804-052131-b12338
46,0 tok/s
How does this benchmark compare on other GPUs?
Same model on different hardware · 1× concurrent · Generation (tok/s)
NVIDIA RTX PRO 6000 Blackwell Workstation Editionrun-20260727-181606-6ea089
378,6 tok/s
NVIDIA GeForce RTX 5090run-20260728-184455-e8b129
377,6 tok/s
3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Editionrun-20260728-184455-9cb45c
343,9 tok/s
NVIDIA GeForce RTX 5070 Tirun-20260728-140953-194348
276,3 tok/s
NVIDIA GeForce RTX 3090 Tirun-20260727-142111-fb08fa
235,8 tok/s
NVIDIA GeForce RTX 5070 Tirun-20260727-032758-19b085
228,2 tok/s
2× NVIDIA RTX A6000run-20260829-102814-741ac2
201,1 tok/s
NVIDIA RTX A6000run-20260826-072502-338126
194,5 tok/s
2× NVIDIA RTX A6000run-20260827-073158-b33795
181,5 tok/s
NVIDIA RTX A6000run-20260826-082947-cac516
153,7 tok/s
NVIDIA RTX A6000run-20260826-183754-3b5658
137,3 tok/s
Intel Arc Pro B70run-20260729-224611-747762
134,3 tok/s
AMD Radeon PRO W7800 48GBrun-20260829-120719-de2775
131,9 tok/s
Tesla V100-PCIE-32GBrun-20260907-133945-5f69da
129,0 tok/s
Hardware
GPU: NVIDIA GeForce RTX 3090 Ti · 24 GB VRAM
CPU: AMD Ryzen 5 5600X 6-Core Processor
RAM: 30 GB
Mainboard: ASUSTeK COMPUTER INC. PRIME A520M-K
Setup
Runtime: vLLM
Quantization: BF16
Model: gemma-4-E2B-it
Configuration
# LLM-Benchmark Konfiguration
# Modell : gemma-4-E2B-it
# Engine : vLLM
# Run-ID : run-20260804-052131-b12338
# GPU : NVIDIA GeForce RTX 3090 Ti
# CPU : AMD Ryzen 5 5600X 6-Core Processor
# RAM : 30 GB
bench@llm-benchmark:~$ /opt/vllm-gemma/venv/bin/python /opt/vllm-gemma/venv/bin/vllm serve google/gemma-4-E2B-it \
--served-model-name gemma-4-E2B-it \
--host 192.168.41.116 \
--port 8000 \
--dtype bfloat16 \
--max-model-len 8192 \
--gpu-memory-utilization 0.90 \
--enforce-eager
| 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. | gemma-4-E2B-it |
| Kontextlaenge?Maximale Anzahl Tokens (Eingabe + erzeugte Ausgabe zusammen), die das Modell pro Anfrage verarbeiten kann. | 8192 |
| Alias?Anzeigename des Modells nach aussen (served model name), unabhaengig vom Dateinamen. | gemma-4-E2B-it |
| Dtype?Zahlenformat der Modellgewichte bei der Berechnung (z.B. auto, float16, bfloat16). 'auto' waehlt automatisch das vom Modell empfohlene Format. | bfloat16 |
| Kontext?Groesse des Kontextfensters in Token. 0 = der beim Training verwendete Kontext des Modells. | 8192 |
| 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'. | 0.90 |
| Enforce-Eager?Schaltet die optimierte Graph-Ausfuehrung (CUDA-/HIP-Graphs) AB und rechnet Schritt fuer Schritt. Startet schneller und spart etwas VRAM, ist im laufenden Betrieb aber meist langsamer als mit Graphs. | aktiv |
Anzeige
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 RTX A6000 1.568,0 tok/s★ NVIDIA GeForce RTX 3090 Ti 1.492,8 tok/s this runNVIDIA GB10 (DGX Spark) 903,3 tok/sIntel Arc Pro B70 826,2 tok/sAMD Radeon PRO W7800 48GB 702,9 tok/sTesla V100-PCIE-32GB 361,8 tok/sNVIDIA GeForce RTX 2060 54,7 tok/s
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 Threadripper PRO 7955WX 16-Cores 1.568,0 tok/sAMD Ryzen 9 8945HX with Radeon Graphics 1.492,8 tok/sNVIDIA Grace 903,3 tok/sAMD Ryzen 9 7945HX with Radeon Graphics 826,2 tok/s★ AMD Ryzen 5 5600X 6-Core Processor 451,1 tok/s this runAMD Ryzen 3 PRO 3200GE w/ Radeon Vega Graphics 361,8 tok/sIntel(R) Xeon(R) CPU E5-4657L v2 @ 2.40GHz 54,7 tok/sIntel(R) Core(TM) i5-7400 CPU @ 3.00GHz 15,3 tok/s
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/sShenzhen Meigao Electronic Equipment Co.,Ltd DRFXI (MotherBoard Series) 826,2 tok/s★ ASUSTeK COMPUTER INC. PRIME A520M-K 451,1 tok/s this runASUSTeK COMPUTER INC. ROG STRIX X570-F GAMING 361,8 tok/sDell Inc. PowerEdge R820 54,7 tok/sASRock H110 Pro BTC+ 15,3 tok/s
DRVby driver
unbekannt 2.491,2 tok/sNVIDIA 590.48.01 / CUDA 13.1 903,3 tok/sIntel 26.18.38308.4 826,2 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
25 W
⚡ TDP
450 W
☁️ External LLM (API)
Electricity0.30 EUR/kWh
Avg power (incl. idle)450 W estimated (TDP)GPU 450 W full load
Avg cost / hourEUR 0.14
Electricity / 1M tokensEUR 0.82
Token / kWh368.08K
Acquisition (system)EUR 1,270 partial priceGPU EUR 999 · CPU EUR 121 · PSU EUR 150
Electricity (2 years)–
TCO (2 years)EUR 3,635
Output tokens (2 years)2.90B
☁️ 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 (25 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-itNVIDIA GeForce RTX 3090 Ti | gemma-4-E2B-itNVIDIA RTX PRO 6000 Blackwell Workstation Edition | gemma-4-E2B-itNVIDIA GeForce RTX 5090 | 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) | – | – | – | – |
NVIDIA GeForce RTX 3090 Ti · this run 46 tok/sNVIDIA RTX PRO 6000 Blackwell Workstation Edition 379 tok/sNVIDIA GeForce RTX 5090 378 tok/s3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition 344 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.