Created bymario-alka.dePowered bygodcore.denoob2claw.detricoma.de
Contributed byMario AlkaGoogle

gemma-4-E4B-it

Performance benchmark · measured on 27.07.2026 08:41

Benchmark-IDrun-20260727-090152-24cf86
Timebench 3 - Kombi (Prefill + Generation)Dense8BRuntime: llama.cppQuantisierung: Q4_K_M
Generation592,13tok/s
Prefill5.662,13tok/s
Time to First Token3.384,50ms
Total duration34,73s
Concurrency5parallel
Ranking in the field
10of 45 systems

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

This run is better than 80 % of all comparable systems.
Generation 592,1 tok/s
+95 % vs Ø 302,9
Prefill 5.662,1 tok/s
+23 % vs Ø 4.604,8
Time to First Token 3.385 ms
-78 % vs Ø 15.587
Distribution in the field2 – 1.743 tok/s
Ø 303 Median Dieser Lauf

Wie schlägt sich dieser Benchmark?

Hardware

GPU: NVIDIA GeForce RTX 5070 Ti · 16 GB VRAM
CPU: AMD Ryzen Threadripper PRO 5975WX 32-Cores
RAM: 247 GB
Mainboard: ASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI

Setup

Runtime: llama.cpp
Quantization: Q4_K_M
Model: gemma-4-E4B-it

Configuration

benchmark-konfiguration — run-20260727-090152-24cf86
# LLM-Benchmark Konfiguration # Modell : gemma-4-E4B-it # Engine : llama.cpp # Run-ID : run-20260727-090152-24cf86 # GPU : NVIDIA GeForce RTX 5070 Ti # CPU : AMD Ryzen Threadripper PRO 5975WX 32-Cores # RAM : 247 GB bench@llm-benchmark:~$ /home/godcore/llama.cpp/build/bin/llama-server \ -m /home/godcore/.cache/huggingface/hub/models--unsloth--gemma-4-E4B-it-GGUF/snapshots/bfc15c382204943c3a8fff0c750b94ae2364d7a3/gemma-4-E4B-it-Q4_K_M.gguf \ --alias gemma-4-E4B-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-E4B-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./home/godcore/.cache/huggingface/hub/models--unsloth--gemma-4-E4B-it-GGUF/snapshots/bfc15c382204943c3a8fff0c750b94ae2364d7a3/gemma-4-E4B-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
np4

All benchmarks of this model To leaderboard

Model comparison

gemma-4-E4B-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

1.3911.0436963480,003.7487.49611.243Prefill (tok/s)Generation (tok/s)NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition - 106,3 tok/s Generation, 2.622 tok/s Prefill, TTFT 3.756 ms (24 Laufe)NVIDIA RTX PRO 6000 B...AMD Radeon 8060S Graphics - 79,7 tok/s Generation, 800 tok/s Prefill, TTFT 12.955 ms (12 Laufe)AMD Radeon 8060S Grap...NVIDIA GeForce RTX 5070 Ti - 1.082,4 tok/s Generation, 9.183 tok/s Prefill, TTFT 2.840 ms (6 Laufe) | DIESER LAUF★ NVIDIA GeForce RTX 50...
★ NVIDIA GeForce RTX 5070 Ti 1.082,4 tok/s this runNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition 106,3 tok/sAMD Radeon 8060S Graphics 79,7 tok/s

CPUby processor

1.3911.0436963480,003.7487.49611.243Prefill (tok/s)Generation (tok/s)AMD Ryzen Threadripper PRO 9965WX 24-Cores - 106,3 tok/s Generation, 2.622 tok/s Prefill, TTFT 3.756 ms (24 Laufe)AMD Ryzen Threadrippe...AMD RYZEN AI MAX+ 395 w/ Radeon 8060S - 79,7 tok/s Generation, 800 tok/s Prefill, TTFT 12.955 ms (12 Laufe)AMD RYZEN AI MAX+ 395...AMD Ryzen Threadripper PRO 5975WX 32-Cores - 1.082,4 tok/s Generation, 9.183 tok/s Prefill, TTFT 2.840 ms (6 Laufe) | DIESER LAUF★ AMD Ryzen Threadrippe...
★ AMD Ryzen Threadripper PRO 5975WX 32-Cores 1.082,4 tok/s this runAMD Ryzen Threadripper PRO 9965WX 24-Cores 106,3 tok/sAMD RYZEN AI MAX+ 395 w/ Radeon 8060S 79,7 tok/s

MBby mainboard

1.3911.0436963480,003.7487.49611.243Prefill (tok/s)Generation (tok/s)ASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE - 106,3 tok/s Generation, 2.622 tok/s Prefill, TTFT 3.756 ms (24 Laufe)ASUSTeK COMPUTER INC....Bosgame AXB35-02 (BeyondMax Series) - 79,7 tok/s Generation, 800 tok/s Prefill, TTFT 12.955 ms (12 Laufe)Bosgame AXB35-02 (Bey...ASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI - 1.082,4 tok/s Generation, 9.183 tok/s Prefill, TTFT 2.840 ms (6 Laufe) | DIESER LAUF★ ASUSTeK COMPUTER INC....
★ ASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI 1.082,4 tok/s this runASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE 106,3 tok/sBosgame AXB35-02 (BeyondMax Series) 79,7 tok/s

ENGby engine

1.2031.1281.05397890305.32110.64115.962Prefill (tok/s)Generation (tok/s)vLLM - 1.022,8 tok/s Generation, 13.200 tok/s Prefill, TTFT 1.096 ms (3 Laufe)vLLMllama.cpp - 1.082,4 tok/s Generation, 2.257 tok/s Prefill, TTFT 6.650 ms (39 Laufe) | DIESER LAUF★ llama.cpp
★ llama.cpp 1.082,4 tok/s this runvLLM 1.022,8 tok/s

DRVby driver

1.1911.1371.0821.0289742.8562.9783.1003.221Prefill (tok/s)Generation (tok/s)unbekannt - 1.082,4 tok/s Generation, 3.039 tok/s Prefill, TTFT 6.254 ms (42 Laufe)unbekannt
unbekannt 1.082,4 tok/s
💰 Economics

Economics of this run

Operating cost, TCO and comparison with the next-best runs of the same model. Methodology →

⚙️ ConfigurationAll metrics and charts below follow these settings – based on a 24-month runtime.
⚡ Electricity price EUR/kWh Save to URL
🖥️ Acquisition EUR Reset
☁️ External LLM (API)
⚙️ System utilization 100 % Idle: 10 W · Full load: 10 W
Electricity0.30 EUR/kWh
Avg power (incl. idle)10 W missingBoard 10 W full load
Avg cost / hourEUR 0.0030
Electricity / 1M tokensEUR 0.0014
Token / kWh213.17M
Acquisition (system)EUR 2,096 missingRAM EUR 1,976 · PSU EUR 120
Electricity (2 years)
TCO (2 years)EUR 2,149
Output tokens (2 years)37.35B
☁️ 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 (10 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 with up to 3 next-best runs of this model (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.