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

Qwen2.5-7B-Instruct

Performance benchmark · measured on 26.09.2026 17:00

Benchmark-IDrun-20260926-150331-53515e
Timebench 3 - Kombi (Prefill + Generation)Dense7BRuntime: llama.cppQuantisierung: Q4_K_M
Generation261,39tok/s
Prefill10.100,85tok/s
Time to First Token2.985,00ms
Total duration84,32s
Concurrency10parallel
Ranking in the field
650of 1301 systems

Performance benchmark · Primary metric: Generation-Speed (tok/s) · 10× concurrent

This run is better than 50 % of all comparable systems.
Generation 261,4 tok/s
-40 % vs Ø 438,1
Prefill 10.100,9 tok/s
+55 % vs Ø 6.502,8
Time to First Token 2.985 ms
-93 % vs Ø 44.915
Distribution in the field0 – 2.491 tok/s
Ø 438 Median Dieser Lauf

Wie schlägt sich dieser Benchmark mit anderen Modellen?

gemma-4-E2B-itNVIDIA GeForce RTX 5090 · run-20260728-184455-e28775
2.491,2 tok/s
gemma-4-E2B-itNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260727-181606-b37ed9
2.414,4 tok/s
Nemotron-3-Nano-4BNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260727-181607-4cca42
2.182,7 tok/s
gemma-4-E2B-it3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-184456-d87c73
2.143,6 tok/s
gpt-oss-20bNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260730-035052-9c966e
1.995,4 tok/s
gpt-oss-20bNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260728-140954-8007c4
1.993,5 tok/s
Nemotron-3-Nano-4B3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-194136-0aed86
1.937,7 tok/s
Nemotron-3-Nano-4B3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-194136-8db0fa
1.911,6 tok/s
Nemotron-3-Nano-4B3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260727-181606-e0eafd
1.897,0 tok/s
gpt-oss-20bNVIDIA GeForce RTX 5090 · run-20260729-032121-7c961d
1.890,7 tok/s
Nemotron-3-Nano-4BNVIDIA GeForce RTX 5090 · run-20260728-194135-2ed9dd
1.878,7 tok/s
gpt-oss-20b3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-034052-26063b
1.835,7 tok/s
Nemotron-3-Nano-4B3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-194136-a77046
1.830,3 tok/s
gpt-oss-20b3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260729-032121-c6e7e0
1.819,5 tok/s
Qwen2.5-7B-Instruct this run2× AMD Radeon RX 7900 XTX · run-20260926-150331-53515e
261,4 tok/s

How does this benchmark compare on other GPUs?

Same model on different hardware · 10× concurrent · Generation (tok/s)

Hardware

GPU: 2x AMD Radeon RX 7900 XTX · 24 GB VRAM
CPU: AMD Ryzen Threadripper PRO 3955WX 16-Cores
RAM: 63 GB
Mainboard: ASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI

Setup

Runtime: llama.cpp
Quantization: Q4_K_M
Model: Qwen2.5-7B-Instruct

Configuration

benchmark-konfiguration — run-20260926-150331-53515e
# LLM-Benchmark Konfiguration # Modell : Qwen2.5-7B-Instruct # Engine : llama.cpp # Run-ID : run-20260926-150331-53515e # GPU : 2x AMD Radeon RX 7900 XTX # CPU : AMD Ryzen Threadripper PRO 3955WX 16-Cores # RAM : 63 GB bench@llm-benchmark:~$ llama-server \ -m Qwen2.5-7B-Instruct-Q4_K_M.gguf \ --alias Qwen2.5-7B-Instruct \ -ngl 999 \ -fa on \ -c 40960 \ -np 10 '(2x' RX 7900 XTX 'ROCm)'
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.Qwen2.5-7B-Instruct
Kontextlaenge?Maximale Anzahl Tokens (Eingabe + erzeugte Ausgabe zusammen), die das Modell pro Anfrage verarbeiten kann.40960
Modellpfad?Pfad zur GGUF-Modelldatei, die geladen und ausgeliefert wird.Qwen2.5-7B-Instruct-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
faon
Kontext?Groesse des Kontextfensters in Token. 0 = der beim Training verwendete Kontext des Modells.40960
np10

All benchmarks of this model To leaderboard

Anzeige
Model comparison

Qwen2.5-7B-Instruct 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

54342230118059,72.0075.0778.14711.216Prefill (tok/s)Generation (tok/s)AMD Radeon AI PRO R9700 - 250,6 tok/s Generation, 6.580 tok/s Prefill, TTFT 2.039 ms (3 Laufe)AMD Radeon AI PRO R97...NVIDIA GeForce RTX 2060 - 160,3 tok/s Generation, 3.649 tok/s Prefill, TTFT 3.816 ms (3 Laufe)NVIDIA GeForce RTX 20...AMD Radeon RX 7900 XTX - 442,0 tok/s Generation, 9.575 tok/s Prefill, TTFT 1.732 ms (6 Laufe) | DIESER LAUF★ AMD Radeon RX 7900 XTX
★ AMD Radeon RX 7900 XTX 442,0 tok/s this runAMD Radeon AI PRO R9700 250,6 tok/sNVIDIA GeForce RTX 2060 160,3 tok/s

CPUby processor

54342230118059,72.0075.0778.14711.216Prefill (tok/s)Generation (tok/s)AMD Ryzen 3 3100 4-Core Processor - 250,6 tok/s Generation, 6.580 tok/s Prefill, TTFT 2.039 ms (3 Laufe)AMD Ryzen 3 3100 4-Co...Intel(R) Xeon(R) CPU E5-4657L v2 @ 2.40GHz - 160,3 tok/s Generation, 3.649 tok/s Prefill, TTFT 3.816 ms (3 Laufe)Intel(R) Xeon(R) CPU ...AMD Ryzen Threadripper PRO 3955WX 16-Cores - 442,0 tok/s Generation, 9.575 tok/s Prefill, TTFT 1.732 ms (6 Laufe) | DIESER LAUF★ AMD Ryzen Threadrippe...
★ AMD Ryzen Threadripper PRO 3955WX 16-Cores 442,0 tok/s this runAMD Ryzen 3 3100 4-Core Processor 250,6 tok/sIntel(R) Xeon(R) CPU E5-4657L v2 @ 2.40GHz 160,3 tok/s

MBby mainboard

54342230118059,72.0075.0778.14711.216Prefill (tok/s)Generation (tok/s)ASUSTeK COMPUTER INC. ROG STRIX X570-F GAMING - 250,6 tok/s Generation, 6.580 tok/s Prefill, TTFT 2.039 ms (3 Laufe)ASUSTeK COMPUTER INC....Dell Inc. PowerEdge R820 - 160,3 tok/s Generation, 3.649 tok/s Prefill, TTFT 3.816 ms (3 Laufe)Dell Inc. PowerEdge R...ASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI - 442,0 tok/s Generation, 9.575 tok/s Prefill, TTFT 1.732 ms (6 Laufe) | DIESER LAUF★ ASUSTeK COMPUTER INC....
★ ASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI 442,0 tok/s this runASUSTeK COMPUTER INC. ROG STRIX X570-F GAMING 250,6 tok/sDell Inc. PowerEdge R820 160,3 tok/s

ENGby engine

5224373522661814.0127.24910.48513.722Prefill (tok/s)Generation (tok/s)vLLM - 442,0 tok/s Generation, 11.912 tok/s Prefill, TTFT 1.633 ms (3 Laufe)vLLMllama.cpp - 261,4 tok/s Generation, 5.822 tok/s Prefill, TTFT 2.562 ms (9 Laufe) | DIESER LAUF★ llama.cpp
vLLM 442,0 tok/s★ llama.cpp 261,4 tok/s this run

DRVby driver

4864644424203986.9047.1987.4927.785Prefill (tok/s)Generation (tok/s)unbekannt - 442,0 tok/s Generation, 7.345 tok/s Prefill, TTFT 2.330 ms (12 Laufe)unbekannt
unbekannt 442,0 tok/s
💰 Economics

Economics of this run

Operating cost, TCO and comparison with the next-best runs of the same model at identical concurrency (10× 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 720 W
☁️ External LLM (API)
Electricity0.30 EUR/kWh
Avg power (incl. idle)720 W estimated (TDP)GPU 710 + Board 10 W full load
Avg cost / hourEUR 0.22
Electricity / 1M tokensEUR 0.23
Token / kWh1.31M
Acquisition (system)EUR 2,782 partial priceGPU EUR 2,098 · RAM EUR 504 · PSU EUR 180
Electricity (2 years)–
TCO (2 years)EUR 6,566
Output tokens (2 years)16.49B
☁️ 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

Qwen2.5-7B-Instruct2x AMD Radeon RX 7900 XTXQwen2.5-7B-Instruct2x AMD Radeon RX 7900 XTXQwen2.5-7B-InstructAMD Radeon AI PRO R9700Qwen2.5-7B-Instruct2x NVIDIA GeForce RTX 2060
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.