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

Step-3.5-Flash

Performance benchmark · measured on 05.08.2026 23:56

Benchmark-IDrun-20260806-094440-0e4a35
Timebench 3 - Kombi (Prefill + Generation)Runtime: llama.cpp
Generation27,45tok/s
Prefill85,67tok/s
Time to First Token171.406,00ms
Total duration943,50s
Concurrency5parallel
Ranking in the field
584of 774 systems

Performance benchmark · Primary metric: Generation-Speed (tok/s) · 5× concurrent

This run is better than 25 % of all comparable systems.
Generation 27,5 tok/s
-91 % vs Ø 300,7
Prefill 85,7 tok/s
-98 % vs Ø 4.088,3
Time to First Token 171.406 ms
+180 % vs Ø 61.233
Distribution in the field0 – 1.349 tok/s
Ø 291 Median Dieser Lauf

Wie schlägt sich dieser Benchmark mit anderen Modellen?

gemma-4-E2B-itNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260727-181606-0adf6d
1.349,3 tok/s
gemma-4-E2B-itNVIDIA GeForce RTX 5090 · run-20260728-184455-5b4937
1.301,9 tok/s
gemma-4-E2B-it3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-184456-4111a9
1.195,7 tok/s
NVIDIA-Nemotron-3-Nano-4BNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260727-181607-30d864
1.188,7 tok/s
NVIDIA-Nemotron-3-Nano-4BNVIDIA GeForce RTX 5090 · run-20260728-194135-5432cd
1.164,8 tok/s
gpt-oss-20bNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260728-140954-9e81f9
1.136,3 tok/s
gpt-oss-20bNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260730-035052-09627f
1.135,0 tok/s
gpt-oss-20bNVIDIA GeForce RTX 5090 · run-20260729-032121-058a31
1.113,5 tok/s
NVIDIA-Nemotron-3-Nano-4B3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-194136-836a91
1.090,6 tok/s
NVIDIA-Nemotron-3-Nano-4B3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260727-181606-8b381e
1.077,4 tok/s
NVIDIA-Nemotron-3-Nano-4B3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-194136-b8a818
1.071,0 tok/s
NVIDIA-Nemotron-3-Nano-4B3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-194136-038e92
1.051,1 tok/s
gpt-oss-20b3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260729-032121-ffb18a
1.015,5 tok/s
NVIDIA-Nemotron-3-Nano-30B-A3BNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260729-032121-305ad6
1.007,8 tok/s
Step-3.5-Flash this runAMD Radeon AI PRO R9700 · run-20260806-094440-0e4a35
27,5 tok/s

How does this benchmark compare on other GPUs?

Same model on different hardware · 5× concurrent · Generation (tok/s)

Hardware

GPU: AMD Radeon AI PRO R9700 · 32 GB VRAM
CPU: AMD Ryzen Threadripper PRO 7955WX 16-Cores
RAM: 184 GB
Mainboard: ASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE

Setup

Runtime: llama.cpp
Quantization: -
Model: Step-3.5-Flash

Configuration

benchmark-konfiguration — run-20260806-094440-0e4a35
# LLM-Benchmark Konfiguration # Modell : Step-3.5-Flash # Engine : llama.cpp # Run-ID : run-20260806-094440-0e4a35 # GPU : AMD Radeon AI PRO R9700 # CPU : AMD Ryzen Threadripper PRO 7955WX 16-Cores # RAM : 184 GB bench@llm-benchmark:~$ /home/godcore/llama.cpp/build/bin/llama-server \ -m /home/godcore/.cache/huggingface/hub/models--stepfun-ai--Step-3.5-Flash-GGUF-Q4_K_S/snapshots/6930492fee9ffd6f13c2e665a94244c60d2d7d83/step3p5_flash_Q4_K_S-00001-of-00012.gguf \ --alias Step-3.5-Flash \ --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.Step-3.5-Flash
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--stepfun-ai--Step-3.5-Flash-GGUF-Q4_K_S/snapshots/6930492fee9ffd6f13c2e665a94244c60d2d7d83/step3p5_flash_Q4_K_S-00001-of-00012.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
Videoreihe und Benchmarkauswertung zur AMD R9700
Model comparison

Step-3.5-Flash 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

9587194792400,009731.9452.918Prefill (tok/s)Generation (tok/s)NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition - 739,1 tok/s Generation, 2.361 tok/s Prefill, TTFT 19.901 ms (12 Laufe)NVIDIA RTX PRO 6000 B...NVIDIA RTX PRO 6000 Blackwell Workstation Edition - 45,5 tok/s Generation, 174 tok/s Prefill, TTFT 129.867 ms (3 Laufe)NVIDIA RTX PRO 6000 B...NVIDIA GeForce RTX 5070 Ti - 30,1 tok/s Generation, 122 tok/s Prefill, TTFT 126.302 ms (3 Laufe)NVIDIA GeForce RTX 50...NVIDIA GeForce RTX 3090 Ti - 17,1 tok/s Generation, 55 tok/s Prefill, TTFT 149.017 ms (3 Laufe)NVIDIA GeForce RTX 30...NVIDIA GeForce RTX 5090 - 13,8 tok/s Generation, 86 tok/s Prefill, TTFT 111.357 ms (3 Laufe)NVIDIA GeForce RTX 50...AMD Radeon AI PRO R9700 - 31,9 tok/s Generation, 123 tok/s Prefill, TTFT 144.454 ms (11 Laufe) | DIESER LAUF★ AMD Radeon AI PRO R97...
NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition 739,1 tok/sNVIDIA RTX PRO 6000 Blackwell Workstation Edition 45,5 tok/s★ AMD Radeon AI PRO R9700 31,9 tok/s this runNVIDIA GeForce RTX 5070 Ti 30,1 tok/sNVIDIA GeForce RTX 3090 Ti 17,1 tok/sNVIDIA GeForce RTX 5090 13,8 tok/s

CPUby processor

9587194792400,009731.9452.918Prefill (tok/s)Generation (tok/s)AMD Ryzen Threadripper PRO 9965WX 24-Cores - 739,1 tok/s Generation, 2.361 tok/s Prefill, TTFT 19.901 ms (12 Laufe)AMD Ryzen Threadrippe...AMD Ryzen 9 9950X 16-Core Processor - 45,5 tok/s Generation, 174 tok/s Prefill, TTFT 129.867 ms (3 Laufe)AMD Ryzen 9 9950X 16-...AMD Ryzen Threadripper PRO 5975WX 32-Cores - 30,1 tok/s Generation, 122 tok/s Prefill, TTFT 126.302 ms (3 Laufe)AMD Ryzen Threadrippe...AMD Ryzen 9 8945HX with Radeon Graphics - 17,1 tok/s Generation, 55 tok/s Prefill, TTFT 149.017 ms (3 Laufe)AMD Ryzen 9 8945HX wi...AMD Ryzen 7 5800X3D 8-Core Processor - 13,8 tok/s Generation, 86 tok/s Prefill, TTFT 111.357 ms (3 Laufe)AMD Ryzen 7 5800X3D 8...AMD Ryzen Threadripper PRO 7955WX 16-Cores - 31,9 tok/s Generation, 123 tok/s Prefill, TTFT 144.454 ms (11 Laufe) | DIESER LAUF★ AMD Ryzen Threadrippe...
AMD Ryzen Threadripper PRO 9965WX 24-Cores 739,1 tok/sAMD Ryzen 9 9950X 16-Core Processor 45,5 tok/s★ AMD Ryzen Threadripper PRO 7955WX 16-Cores 31,9 tok/s this runAMD Ryzen Threadripper PRO 5975WX 32-Cores 30,1 tok/sAMD Ryzen 9 8945HX with Radeon Graphics 17,1 tok/sAMD Ryzen 7 5800X3D 8-Core Processor 13,8 tok/s

MBby mainboard

9587194792400,005301.0601.590Prefill (tok/s)Generation (tok/s)ASUSTeK COMPUTER INC. ProArt X870E-CREATOR WIFI - 45,5 tok/s Generation, 174 tok/s Prefill, TTFT 129.867 ms (3 Laufe)ASUSTeK COMPUTER INC....ASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI - 30,1 tok/s Generation, 122 tok/s Prefill, TTFT 126.302 ms (3 Laufe)ASUSTeK COMPUTER INC....Meigao Innovation Technology (Shen Zhen) Co., Ltd DRFXL (MotherBoard Series) - 17,1 tok/s Generation, 55 tok/s Prefill, TTFT 149.017 ms (3 Laufe)Meigao Innovation Tec...ASUSTeK COMPUTER INC. ROG STRIX B550-A GAMING - 13,8 tok/s Generation, 86 tok/s Prefill, TTFT 111.357 ms (3 Laufe)ASUSTeK COMPUTER INC....ASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE - 739,1 tok/s Generation, 1.291 tok/s Prefill, TTFT 79.470 ms (23 Laufe) | DIESER LAUF★ ASUSTeK COMPUTER INC....
★ ASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE 739,1 tok/s this runASUSTeK COMPUTER INC. ProArt X870E-CREATOR WIFI 45,5 tok/sASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI 30,1 tok/sMeigao Innovation Technology (Shen Zhen) Co., Ltd DRFXL (MotherBoard Series) 17,1 tok/sASUSTeK COMPUTER INC. ROG STRIX B550-A GAMING 13,8 tok/s

ENGby engine

813776739702665832868903939Prefill (tok/s)Generation (tok/s)llama.cpp - 739,1 tok/s Generation, 886 tok/s Prefill, TTFT 96.498 ms (35 Laufe) | DIESER LAUF★ llama.cpp
★ llama.cpp 739,1 tok/s this run

DRVby driver

813776739702665832868903939Prefill (tok/s)Generation (tok/s)unbekannt - 739,1 tok/s Generation, 886 tok/s Prefill, TTFT 96.498 ms (35 Laufe)unbekannt
unbekannt 739,1 tok/s
💰 Economics

Economics of this run

Operating cost, TCO and comparison with the next-best runs of the same model at identical concurrency (5× 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 85 W
⚡ TDP 371 W
☁️ External LLM (API)
Electricity0.30 EUR/kWh
Avg power (incl. idle)371 W estimated (TDP)GPU 300 + CPU 46 + Board 25 W full load
Avg cost / hourEUR 0.11
Electricity / 1M tokensEUR 1.13
Token / kWh266.36K
Acquisition (system)EUR 6,824 full priceGPU EUR 1,400 · CPU EUR 1,399 · Board EUR 1,299 · RAM EUR 2,576 · PSU EUR 150
Electricity (2 years)
TCO (2 years)EUR 8,774
Output tokens (2 years)1.73B
☁️ 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 (85 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

Step-3.5-FlashAMD Radeon AI PRO R9700Step-3.5-Flash3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionStep-3.5-Flash3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionStep-3.5-Flash3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition
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.