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

Gemma-4-26B-A4B-it

Performance benchmark · measured on 28.07.2026 23:46

Benchmark-IDrun-20260729-032121-c5db85
Timebench 3 - Kombi (Prefill + Generation)MoE26BRuntime: llama.cppQuantisierung: UD-Q4_K_XL
Generation189,07tok/s
Prefill3.110,21tok/s
Time to First Token638,00ms
Total duration12,11s
Concurrency1parallel
Ranking in the field
106of 352 systems

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

This run is better than 70 % of all comparable systems.
Generation 189,1 tok/s
+62 % vs Ø 116,6
Prefill 3.110,2 tok/s
0 % vs Ø 3.123,6
Time to First Token 638 ms
-93 % vs Ø 9.722
Distribution in the field0 – 405 tok/s
Ø 117 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
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
Nemotron-Cascade-2-30B-A3BNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260729-032120-4e3476
356,8 tok/s
NVIDIA-Nemotron-3-Nano-4B3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260727-181606-a2f36d
356,7 tok/s
NVIDIA-Nemotron-3-Nano-4B3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-194136-1be57b
355,6 tok/s
Gemma-4-26B-A4B-it this runNVIDIA GeForce RTX 5070 Ti · run-20260729-032121-c5db85
189,1 tok/s

How does this benchmark compare on other GPUs?

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

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: UD-Q4_K_XL
Model: Gemma-4-26B-A4B-it

Configuration

benchmark-konfiguration — run-20260729-032121-c5db85
# LLM-Benchmark Konfiguration # Modell : Gemma-4-26B-A4B-it # Engine : llama.cpp # Run-ID : run-20260729-032121-c5db85 # 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-4-26B-A4B-it-qat-GGUF/snapshots/7b92b5b28818151e8669af2e45e88d6086f490dd/gemma-4-26B-A4B-it-qat-UD-Q4_K_XL.gguf \ --alias Gemma-4-26B-A4B-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-4-26B-A4B-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-4-26B-A4B-it-qat-GGUF/snapshots/7b92b5b28818151e8669af2e45e88d6086f490dd/gemma-4-26B-A4B-it-qat-UD-Q4_K_XL.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

Gemma-4-26B-A4B-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

1.2889666443220,0431.4792.9154.350Prefill (tok/s)Generation (tok/s)NVIDIA GeForce RTX 3090 Ti - 926,2 tok/s Generation, 3.620 tok/s Prefill, TTFT 6.478 ms (3 Laufe)NVIDIA GeForce RTX 30...AMD Radeon 8060S Graphics - 119,5 tok/s Generation, 773 tok/s Prefill, TTFT 6.904 ms (5 Laufe)AMD Radeon 8060S Grap...NVIDIA GeForce RTX 5070 Ti - 1.009,0 tok/s Generation, 3.578 tok/s Prefill, TTFT 5.953 ms (3 Laufe) | DIESER LAUF★ NVIDIA GeForce RTX 50...
★ NVIDIA GeForce RTX 5070 Ti 1.009,0 tok/s this runNVIDIA GeForce RTX 3090 Ti 926,2 tok/sAMD Radeon 8060S Graphics 119,5 tok/s

CPUby processor

1.2889666443220,0431.4792.9154.350Prefill (tok/s)Generation (tok/s)AMD Ryzen 9 8945HX with Radeon Graphics - 926,2 tok/s Generation, 3.620 tok/s Prefill, TTFT 6.478 ms (3 Laufe)AMD Ryzen 9 8945HX wi...AMD RYZEN AI MAX+ 395 w/ Radeon 8060S - 119,5 tok/s Generation, 773 tok/s Prefill, TTFT 6.904 ms (5 Laufe)AMD RYZEN AI MAX+ 395...AMD Ryzen Threadripper PRO 5975WX 32-Cores - 1.009,0 tok/s Generation, 3.578 tok/s Prefill, TTFT 5.953 ms (3 Laufe) | DIESER LAUF★ AMD Ryzen Threadrippe...
★ AMD Ryzen Threadripper PRO 5975WX 32-Cores 1.009,0 tok/s this runAMD Ryzen 9 8945HX with Radeon Graphics 926,2 tok/sAMD RYZEN AI MAX+ 395 w/ Radeon 8060S 119,5 tok/s

MBby mainboard

1.2889666443220,0431.4792.9154.350Prefill (tok/s)Generation (tok/s)Meigao Innovation Technology (Shen Zhen) Co., Ltd DRFXL (MotherBoard Series) - 926,2 tok/s Generation, 3.620 tok/s Prefill, TTFT 6.478 ms (3 Laufe)Meigao Innovation Tec...Bosgame AXB35-02 (BeyondMax Series) - 119,5 tok/s Generation, 773 tok/s Prefill, TTFT 6.904 ms (5 Laufe)Bosgame AXB35-02 (Bey...ASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI - 1.009,0 tok/s Generation, 3.578 tok/s Prefill, TTFT 5.953 ms (3 Laufe) | DIESER LAUF★ ASUSTeK COMPUTER INC....
★ ASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI 1.009,0 tok/s this runMeigao Innovation Technology (Shen Zhen) Co., Ltd DRFXL (MotherBoard Series) 926,2 tok/sBosgame AXB35-02 (BeyondMax Series) 119,5 tok/s

ENGby engine

1.1101.0591.0099599082.1762.2682.3612.454Prefill (tok/s)Generation (tok/s)llama.cpp - 1.009,0 tok/s Generation, 2.315 tok/s Prefill, TTFT 6.528 ms (11 Laufe) | DIESER LAUF★ llama.cpp
★ llama.cpp 1.009,0 tok/s this run

DRVby driver

1.1101.0591.0099599082.1762.2682.3612.454Prefill (tok/s)Generation (tok/s)unbekannt - 1.009,0 tok/s Generation, 2.315 tok/s Prefill, TTFT 6.528 ms (11 Laufe)unbekannt
unbekannt 1.009,0 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 10 W
⚡ TDP 310 W
☁️ External LLM (API)
Electricity0.30 EUR/kWh
Avg power (incl. idle)310 W estimated (TDP)GPU 300 + Board 10 W full load
Avg cost / hourEUR 0.093
Electricity / 1M tokensEUR 0.14
Token / kWh2.20M
Acquisition (system)EUR 2,126 missingRAM EUR 1,976 · PSU EUR 150
Electricity (2 years)
TCO (2 years)EUR 3,755
Output tokens (2 years)11.93B
☁️ 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 (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 vs. API – economics per benchmark

Gemma-4-26B-A4B-itNVIDIA GeForce RTX 5070 TiGemma-4-26B-A4B-itNVIDIA GeForce RTX 3090 TiGemma-4-26B-A4B-itAMD Radeon 8060S GraphicsGemma-4-26B-A4B-itAMD 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.