Created bymario-alka.dePowered bynoob2claw.detricoma.de
Contributed byMario AlkaQwen (Alibaba)

Qwen3-4B

Performance benchmark · measured on 12.11.2026 15:27

Benchmark-IDrun-20261112-144801-70f324
Timebench 3 - Kombi (Prefill + Generation)Dense4BRuntime: lmstudio-mlxQuantisierung: MLX 4bit
Generation48,34tok/s
Prefill1.926,77tok/s
Time to First Token1.387,70ms
Total duration71,00s
Concurrency1parallel
Ranking in the field
990of 1927 systems

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

This run is better than 49 % of all comparable systems.
Generation 48,3 tok/s
-38 % vs Ø 77,5
Prefill 1.926,8 tok/s
-32 % vs Ø 2.844,6
Time to First Token 1.388 ms
-94 % vs Ø 23.755
Distribution in the field0 – 471 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: Keine GPU (CPU-only)
CPU: Apple M6 (unified CPU/GPU SoC, ARM64, AdvSIMD)
RAM: 24 GB
Mainboard: Apple Mac mini

Setup

Runtime: lmstudio-mlx
Quantization: MLX 4bit
Model: Qwen3-4B

Configuration

benchmark-konfiguration — run-20261112-144801-70f324
# LLM-Benchmark Konfiguration # Modell : Qwen3-4B # Engine : lmstudio-mlx # Run-ID : run-20261112-144801-70f324 # GPU : Keine GPU (CPU-only) # CPU : Apple M6 (unified CPU/GPU SoC, ARM64, AdvSIMD) # RAM : 24 GB bench@llm-benchmark:~$ lmstudio-mlx \ --model qwen3-4b
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.lmstudio-mlx
Kontextlaenge?Maximale Anzahl Tokens (Eingabe + erzeugte Ausgabe zusammen), die das Modell pro Anfrage verarbeiten kann.8192
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

4133102071030,003.9327.86511.797Prefill (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, 9.607 tok/s Prefill, TTFT 1.584 ms (9 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...CPU-only - 48,3 tok/s Generation, 641 tok/s Prefill, TTFT 225.381 ms (5 Laufe) | DIESER LAUF★ CPU-only
AMD Radeon AI PRO R9700 325,2 tok/sAMD Radeon RX 7900 XTX 320,4 tok/sAMD Oberon (PlayStation 5) 101,5 tok/s★ CPU-only 48,3 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...Intel(R) Xeon(R) CPU E5-4657L v2 @ 2.40GHz - 294,3 tok/s Generation, 6.880 tok/s Prefill, TTFT 1.765 ms (3 Laufe)Intel(R) Xeon(R) CPU ...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...Apple M6 (unified CPU/GPU SoC) - 41,8 tok/s Generation, 1.253 tok/s Prefill, TTFT 2.021 ms (1 Lauf)Apple M6 (unified CPU...AMD Jaguar 8-Core (PlayStation 4) - 1,3 tok/s Generation, 4 tok/s Prefill, TTFT 474.264 ms (1 Lauf)AMD Jaguar 8-Core (Pl...4 x Cortex-A78C + 4 x Cortex-A55 - 0,9 tok/s Generation, 11 tok/s Prefill, TTFT 324.617 ms (2 Laufe)4 x Cortex-A78C + 4 x...Apple M6 (unified CPU/GPU SoC, ARM64, AdvSIMD) - 48,3 tok/s Generation, 1.927 tok/s Prefill, TTFT 1.388 ms (1 Lauf) | DIESER LAUF★ Apple M6 (unified CPU...
AMD Ryzen 3 3100 4-Core Processor 325,2 tok/sAMD Ryzen Threadripper PRO 3955WX 16-Cores 320,4 tok/sIntel(R) Xeon(R) CPU E5-4657L v2 @ 2.40GHz 294,3 tok/sAMD Zen2 8-Core (PlayStation 5) 101,5 tok/s★ Apple M6 (unified CPU/GPU SoC, ARM64, AdvSIMD) 48,3 tok/s this runApple M6 (unified CPU/GPU SoC) 41,8 tok/sAMD Jaguar 8-Core (PlayStation 4) 1,3 tok/s4 x Cortex-A78C + 4 x Cortex-A55 0,9 tok/s

MBby mainboard

4133102071030,004.5159.03013.545Prefill (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....Dell Inc. PowerEdge R820 - 294,3 tok/s Generation, 6.880 tok/s Prefill, TTFT 1.765 ms (3 Laufe)Dell Inc. PowerEdge R...unbekannt - 101,5 tok/s Generation, 318 tok/s Prefill, TTFT 228.383 ms (5 Laufe)unbekanntApple Mac mini - 48,3 tok/s Generation, 1.590 tok/s Prefill, TTFT 1.704 ms (2 Laufe) | DIESER LAUF★ Apple Mac mini
ASUSTeK COMPUTER INC. ROG STRIX X570-F GAMING 325,2 tok/sASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI 320,4 tok/sDell Inc. PowerEdge R820 294,3 tok/sunbekannt 101,5 tok/s★ Apple Mac mini 48,3 tok/s this run

ENGby engine

4143112071040,005.72011.44017.160Prefill (tok/s)Generation (tok/s)llama.cpp - 325,2 tok/s Generation, 4.826 tok/s Prefill, TTFT 82.646 ms (14 Laufe)llama.cppvLLM - 320,4 tok/s Generation, 14.021 tok/s Prefill, TTFT 1.458 ms (3 Laufe)vLLMOllama - 41,8 tok/s Generation, 1.253 tok/s Prefill, TTFT 2.021 ms (1 Lauf)Ollamalmstudio-mlx - 48,3 tok/s Generation, 1.927 tok/s Prefill, TTFT 1.388 ms (1 Lauf) | DIESER LAUF★ lmstudio-mlx
llama.cpp 325,2 tok/svLLM 320,4 tok/s★ lmstudio-mlx 48,3 tok/s this runOllama 41,8 tok/s

DRVby driver

3583423253092935.5815.8196.0566.294Prefill (tok/s)Generation (tok/s)unbekannt - 325,2 tok/s Generation, 5.937 tok/s Prefill, TTFT 61.307 ms (19 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)– partial price
Electricity (2 years)–
TCO (2 years)–
Output tokens (2 years)3.05B
☁️ 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

Qwen3-4BKeine GPU (CPU-only)Qwen3-4B2x AMD Radeon RX 7900 XTXQwen3-4BAMD Oberon (PlayStation 5)Qwen3-4B2x AMD Radeon RX 7900 XTX
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.