Created bymario-alka.dePowered bygodcore.denoob2claw.detricoma.de
Contributed byMario AlkaQwen (Alibaba)

Qwen3-4B

Performance benchmark · measured on 24.09.2026 16:44

Benchmark-IDrun-20261003-073242-6faa20
Timebench 3 - Kombi (Prefill + Generation)Dense4BRuntime: llama.cppQuantisierung: Q4_K_M
Generation1,25tok/s
Prefill3,98tok/s
Time to First Token474.264,00ms
Total duration1.200,00s
Concurrency1parallel
Ranking in the field
1643of 1704 systems

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

This run is better than 4 % of all comparable systems.
Generation 1,3 tok/s
-98 % vs Ø 78,3
Prefill 4,0 tok/s
-100 % vs Ø 2.873,1
Time to First Token 474.264 ms
+1.742 % vs Ø 25.743
Distribution in the field0 – 405 tok/s
Ø 78 Median Dieser Lauf

Wie schlägt sich dieser Benchmark mit anderen Modellen?

How does this benchmark compare on other GPUs?

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

Hardware

GPU: AMD Liverpool (PlayStation 4)
CPU: AMD Jaguar 8-Core (PlayStation 4)
RAM: 5 GB

Setup

Runtime: llama.cpp
Quantization: Q4_K_M
Model: Qwen3-4B

Configuration

benchmark-konfiguration — run-20261003-073242-6faa20
# LLM-Benchmark Konfiguration # Modell : Qwen3-4B # Engine : llama.cpp # Run-ID : run-20261003-073242-6faa20 # GPU : AMD Liverpool (PlayStation 4) # CPU : AMD Jaguar 8-Core (PlayStation 4) # RAM : 5 GB bench@llm-benchmark:~$ llama-server \ -m Qwen3-4B-Q4_K_M.gguf \ --alias Qwen3-4B \ -c 4096 \ -np 1 \ -t 6 '(PS4' AMD Jaguar 8-Core 'CPU-only)'
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.Qwen3-4B
Kontextlaenge?Maximale Anzahl Tokens (Eingabe + erzeugte Ausgabe zusammen), die das Modell pro Anfrage verarbeiten kann.4096
Modellpfad?Pfad zur GGUF-Modelldatei, die geladen und ausgeliefert wird.Qwen3-4B-Q4_K_M.gguf
Kontext?Groesse des Kontextfensters in Token. 0 = der beim Training verwendete Kontext des Modells.4096
np1
Threads?Anzahl CPU-Threads fuer die Token-Generierung (Decode).6
execution_typelocal

All benchmarks of this model To leaderboard

Anzeige
Model comparison

Qwen3-4B 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

4233172111060,004.5349.06813.602Prefill (tok/s)Generation (tok/s)AMD Radeon AI PRO R9700 - 325,2 tok/s Generation, 7.193 tok/s Prefill, TTFT 1.751 ms (3 Laufe)AMD Radeon AI PRO R97...AMD Radeon RX 7900 XTX - 320,4 tok/s Generation, 10.970 tok/s Prefill, TTFT 1.493 ms (6 Laufe)AMD Radeon RX 7900 XTXAMD Oberon (PlayStation 5) - 101,5 tok/s Generation, 782 tok/s Prefill, TTFT 9.208 ms (2 Laufe)AMD Oberon (PlayStati...AMD Liverpool (PlayStation 4) - 1,3 tok/s Generation, 4 tok/s Prefill, TTFT 474.264 ms (1 Lauf) | DIESER LAUF★ AMD Liverpool (PlaySt...
AMD Radeon AI PRO R9700 325,2 tok/sAMD Radeon RX 7900 XTX 320,4 tok/sAMD Oberon (PlayStation 5) 101,5 tok/s★ AMD Liverpool (PlayStation 4) 1,3 tok/s this run

CPUby processor

4233172111060,004.5349.06813.602Prefill (tok/s)Generation (tok/s)AMD Ryzen 3 3100 4-Core Processor - 325,2 tok/s Generation, 7.193 tok/s Prefill, TTFT 1.751 ms (3 Laufe)AMD Ryzen 3 3100 4-Co...AMD Ryzen Threadripper PRO 3955WX 16-Cores - 320,4 tok/s Generation, 10.970 tok/s Prefill, TTFT 1.493 ms (6 Laufe)AMD Ryzen Threadrippe...AMD Zen2 8-Core (PlayStation 5) - 101,5 tok/s Generation, 782 tok/s Prefill, TTFT 9.208 ms (2 Laufe)AMD Zen2 8-Core (Play...AMD Jaguar 8-Core (PlayStation 4) - 1,3 tok/s Generation, 4 tok/s Prefill, TTFT 474.264 ms (1 Lauf) | DIESER LAUF★ AMD Jaguar 8-Core (Pl...
AMD Ryzen 3 3100 4-Core Processor 325,2 tok/sAMD Ryzen Threadripper PRO 3955WX 16-Cores 320,4 tok/sAMD Zen2 8-Core (PlayStation 5) 101,5 tok/s★ AMD Jaguar 8-Core (PlayStation 4) 1,3 tok/s this run

MBby mainboard

40330821311924,204.5039.00613.508Prefill (tok/s)Generation (tok/s)ASUSTeK COMPUTER INC. ROG STRIX X570-F GAMING - 325,2 tok/s Generation, 7.193 tok/s Prefill, TTFT 1.751 ms (3 Laufe)ASUSTeK COMPUTER INC....ASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI - 320,4 tok/s Generation, 10.970 tok/s Prefill, TTFT 1.493 ms (6 Laufe)ASUSTeK COMPUTER INC....unbekannt - 101,5 tok/s Generation, 523 tok/s Prefill, TTFT 164.227 ms (3 Laufe)unbekannt
ASUSTeK COMPUTER INC. ROG STRIX X570-F GAMING 325,2 tok/sASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI 320,4 tok/sunbekannt 101,5 tok/s

ENGby engine

3593413233052872.7857.33911.89316.448Prefill (tok/s)Generation (tok/s)vLLM - 320,4 tok/s Generation, 14.021 tok/s Prefill, TTFT 1.458 ms (3 Laufe)vLLMllama.cpp - 325,2 tok/s Generation, 5.212 tok/s Prefill, TTFT 55.836 ms (9 Laufe) | DIESER LAUF★ llama.cpp
★ llama.cpp 325,2 tok/s this runvLLM 320,4 tok/s

DRVby driver

3583423253092936.9697.2667.5627.859Prefill (tok/s)Generation (tok/s)unbekannt - 325,2 tok/s Generation, 7.414 tok/s Prefill, TTFT 42.241 ms (12 Laufe)unbekannt
unbekannt 325,2 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 %
🔌 Idle 0 W
⚡ TDP 10 W
☁️ External LLM (API)
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)78.84M
☁️ 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

Qwen3-4BAMD Liverpool (PlayStation 4)Qwen3-4BAMD Oberon (PlayStation 5)Qwen3-4B2x AMD Radeon RX 7900 XTXQwen3-4BAMD Radeon AI PRO R9700
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.