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

gpt-oss-120b

Performance benchmark · measured on 27.07.2026 20:59

Benchmark-IDrun-20260728-034051-d52467
Timebench 3 - Kombi (Prefill + Generation)MoE120BRuntime: llama.cppQuantisierung: Q4_K_M
Generation1.229,72tok/s
Prefill9.543,29tok/s
Time to First Token8.770,50ms
Total duration52,95s
Concurrency10parallel
Ranking in the field
4of 19 systems

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

This run is better than 83 % of all comparable systems.
Generation 1.229,7 tok/s
+142 % vs Ø 509,0
Prefill 9.543,3 tok/s
+54 % vs Ø 6.208,2
Time to First Token 8.771 ms
-53 % vs Ø 18.762
Distribution in the field55 – 1.897 tok/s
Ø 509 Median Dieser Lauf

Wie schlägt sich dieser Benchmark mit anderen Modellen?

NVIDIA-Nemotron-3-Nano-4B3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260727-181606-e0eafd
1.897,0 tok/s
gpt-oss-20b3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-034052-26063b
1.835,7 tok/s
gpt-oss-20b3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260724-082940-10761f
1.238,3 tok/s
gpt-oss-120b this run3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-034051-d52467
1.229,7 tok/s
GLM-4.5-Air3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260727-193100-9a2d59
615,3 tok/s
Devstral-Small-25073× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260727-181606-59af91
572,2 tok/s
gemma-4-31B-it3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260727-193100-f3344f
397,5 tok/s
gemma-4-31B-it3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260727-181607-1949e0
394,5 tok/s
Laguna-M.13× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-034053-212933
337,6 tok/s
gemma-4-E2B-it3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260727-155936-b597f5
266,1 tok/s
gemma-4-E4B-it3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260727-181605-64c902
137,2 tok/s
command-a-reasoning-08-20253× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260727-181606-ca0b50
134,9 tok/s
gpt-oss-20b3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260724-082952-499c95
109,1 tok/s
gpt-oss-20b3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260725-031737-ac6fbc
90,4 tok/s

How does this benchmark compare on other GPUs?

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

Hardware

GPU: 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 96 GB VRAM
CPU: AMD Ryzen Threadripper PRO 9965WX 24-Cores
RAM: 125 GB
Mainboard: ASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE

Setup

Runtime: llama.cpp
Quantization: Q4_K_M
Model: gpt-oss-120b

Configuration

benchmark-konfiguration — run-20260728-034051-d52467
# LLM-Benchmark Konfiguration # Modell : gpt-oss-120b # Engine : llama.cpp # Run-ID : run-20260728-034051-d52467 # GPU : 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition # CPU : AMD Ryzen Threadripper PRO 9965WX 24-Cores # RAM : 125 GB bench@llm-benchmark:~$ /home/godcore/llama.cpp/build/bin/llama-server \ -m /home/godcore/.cache/huggingface/hub/models--unsloth--gpt-oss-120b-GGUF/snapshots/ff1a82da6ad466e32284fa3d2b86694db3204789/Q4_K_M/gpt-oss-120b-Q4_K_M-00001-of-00002.gguf \ --alias gpt-oss-120b \ --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.gpt-oss-120b
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--gpt-oss-120b-GGUF/snapshots/ff1a82da6ad466e32284fa3d2b86694db3204789/Q4_K_M/gpt-oss-120b-Q4_K_M-00001-of-00002.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

gpt-oss-120b 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

1.5981.1987993990,003.2186.4369.653Prefill (tok/s)Generation (tok/s)AMD Radeon AI PRO R9700 - 241,4 tok/s Generation, 2.947 tok/s Prefill, TTFT 5.020 ms (2 Laufe)AMD Radeon AI PRO R97...NVIDIA GeForce RTX 3090 Ti - 85,3 tok/s Generation, 238 tok/s Prefill, TTFT 79.784 ms (3 Laufe)NVIDIA GeForce RTX 30...NVIDIA GeForce RTX 5070 Ti - 60,1 tok/s Generation, 195 tok/s Prefill, TTFT 69.936 ms (3 Laufe)NVIDIA GeForce RTX 50...NVIDIA GeForce RTX 5090 - 13,3 tok/s Generation, 369 tok/s Prefill, TTFT 16.219 ms (2 Laufe)NVIDIA GeForce RTX 50...CPU-only - 5,3 tok/s Generation, 42 tok/s Prefill, TTFT 165.568 ms (3 Laufe)CPU-onlyNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition - 1.229,7 tok/s Generation, 7.791 tok/s Prefill, TTFT 3.978 ms (3 Laufe) | DIESER LAUF★ NVIDIA RTX PRO 6000 B...
★ NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition 1.229,7 tok/s this runAMD Radeon AI PRO R9700 241,4 tok/sNVIDIA GeForce RTX 3090 Ti 85,3 tok/sNVIDIA GeForce RTX 5070 Ti 60,1 tok/sNVIDIA GeForce RTX 5090 13,3 tok/sCPU-only 5,3 tok/s

CPUby processor

1.5981.1987993990,003.2186.4369.653Prefill (tok/s)Generation (tok/s)AMD Ryzen Threadripper PRO 7955WX 16-Cores - 241,4 tok/s Generation, 2.947 tok/s Prefill, TTFT 5.020 ms (2 Laufe)AMD Ryzen Threadrippe...AMD Ryzen 9 8945HX with Radeon Graphics - 85,3 tok/s Generation, 238 tok/s Prefill, TTFT 79.784 ms (3 Laufe)AMD Ryzen 9 8945HX wi...AMD Ryzen Threadripper PRO 5975WX 32-Cores - 60,1 tok/s Generation, 195 tok/s Prefill, TTFT 69.936 ms (3 Laufe)AMD Ryzen Threadrippe...AMD Ryzen 7 5800X3D 8-Core Processor - 13,3 tok/s Generation, 369 tok/s Prefill, TTFT 16.219 ms (2 Laufe)AMD Ryzen 7 5800X3D 8...Intel(R) Xeon(R) CPU E5-4620 v2 @ 2.60GHz - 5,3 tok/s Generation, 42 tok/s Prefill, TTFT 165.568 ms (3 Laufe)Intel(R) Xeon(R) CPU ...AMD Ryzen Threadripper PRO 9965WX 24-Cores - 1.229,7 tok/s Generation, 7.791 tok/s Prefill, TTFT 3.978 ms (3 Laufe) | DIESER LAUF★ AMD Ryzen Threadrippe...
★ AMD Ryzen Threadripper PRO 9965WX 24-Cores 1.229,7 tok/s this runAMD Ryzen Threadripper PRO 7955WX 16-Cores 241,4 tok/sAMD Ryzen 9 8945HX with Radeon Graphics 85,3 tok/sAMD Ryzen Threadripper PRO 5975WX 32-Cores 60,1 tok/sAMD Ryzen 7 5800X3D 8-Core Processor 13,3 tok/sIntel(R) Xeon(R) CPU E5-4620 v2 @ 2.60GHz 5,3 tok/s

MBby mainboard

1.5981.1987993990,002.4174.8347.251Prefill (tok/s)Generation (tok/s)Meigao Innovation Technology (Shen Zhen) Co., Ltd DRFXL (MotherBoard Series) - 85,3 tok/s Generation, 238 tok/s Prefill, TTFT 79.784 ms (3 Laufe)Meigao Innovation Tec...ASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI - 60,1 tok/s Generation, 195 tok/s Prefill, TTFT 69.936 ms (3 Laufe)ASUSTeK COMPUTER INC....ASUSTeK COMPUTER INC. ROG STRIX B550-A GAMING - 13,3 tok/s Generation, 369 tok/s Prefill, TTFT 16.219 ms (2 Laufe)ASUSTeK COMPUTER INC....Dell Inc. PowerEdge R820 - 5,3 tok/s Generation, 42 tok/s Prefill, TTFT 165.568 ms (3 Laufe)Dell Inc. PowerEdge R...ASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE - 1.229,7 tok/s Generation, 5.854 tok/s Prefill, TTFT 4.395 ms (5 Laufe) | DIESER LAUF★ ASUSTeK COMPUTER INC....
★ ASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE 1.229,7 tok/s this runMeigao Innovation Technology (Shen Zhen) Co., Ltd DRFXL (MotherBoard Series) 85,3 tok/sASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI 60,1 tok/sASUSTeK COMPUTER INC. ROG STRIX B550-A GAMING 13,3 tok/sDell Inc. PowerEdge R820 5,3 tok/s

ENGby engine

1.5961.1977983990,008841.7682.652Prefill (tok/s)Generation (tok/s)vLLM - 13,3 tok/s Generation, 369 tok/s Prefill, TTFT 16.219 ms (2 Laufe)vLLMllama.cpp - 1.229,7 tok/s Generation, 2.192 tok/s Prefill, TTFT 69.131 ms (14 Laufe) | DIESER LAUF★ llama.cpp
★ llama.cpp 1.229,7 tok/s this runvLLM 13,3 tok/s

DRVby driver

1.3531.2911.2301.1681.1071.8471.9252.0042.082Prefill (tok/s)Generation (tok/s)unbekannt - 1.229,7 tok/s Generation, 1.964 tok/s Prefill, TTFT 62.517 ms (16 Laufe)unbekannt
unbekannt 1.229,7 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 165 W
⚡ TDP 983 W
☁️ External LLM (API)
Electricity0.30 EUR/kWh
Avg power (incl. idle)983 W estimated (TDP)GPU 900 + CPU 57 + Board 25 W full load
Avg cost / hourEUR 0.29
Electricity / 1M tokensEUR 0.067
Token / kWh4.51M
Acquisition (system)EUR 45,748 full priceGPU EUR 39,000 · CPU EUR 3,499 · Board EUR 1,299 · RAM EUR 1,750 · PSU EUR 200
Electricity (2 years)
TCO (2 years)EUR 50,912
Output tokens (2 years)77.56B
☁️ 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 (165 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

gpt-oss-120b3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Editiongpt-oss-120b3x AMD Radeon AI PRO R9700gpt-oss-120bNVIDIA GeForce RTX 3090 Tigpt-oss-120bNVIDIA GeForce RTX 5070 Ti
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.