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

Ornith-1.0-35B

Performance benchmark · measured on 04.08.2026 10:45

Benchmark-IDrun-20260805-053411-dc2b72
Timebench 3 - Kombi (Prefill + Generation)Dense35BRuntime: llama.cppQuantisierung: Q4_K_M
Generation62,81tok/s
Prefill238,75tok/s
Time to First Token56.817,50ms
Total duration373,72s
Concurrency5parallel
Ranking in the field
475of 704 systems

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

This run is better than 33 % of all comparable systems.
Generation 62,8 tok/s
-81 % vs Ø 323,5
Prefill 238,8 tok/s
-95 % vs Ø 4.464,5
Time to First Token 56.818 ms
+10 % vs Ø 51.689
Distribution in the field0 – 1.349 tok/s
Ø 315 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
Ornith-1.0-35B this runAMD Radeon AI PRO R9700 · run-20260805-053411-dc2b72
62,8 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: Q4_K_M
Model: Ornith-1.0-35B

Configuration

benchmark-konfiguration — run-20260805-053411-dc2b72
# LLM-Benchmark Konfiguration # Modell : Ornith-1.0-35B # Engine : llama.cpp # Run-ID : run-20260805-053411-dc2b72 # 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--deepreinforce-ai--Ornith-1.0-35B-GGUF/snapshots/383064f72a1ef3087b779f268d3ca117eb989aac/ornith-1.0-35b-Q4_K_M.gguf \ --alias Ornith-1.0-35B \ --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.Ornith-1.0-35B
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--deepreinforce-ai--Ornith-1.0-35B-GGUF/snapshots/383064f72a1ef3087b779f268d3ca117eb989aac/ornith-1.0-35b-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
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

Ornith-1.0-35B 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

2.3181.7391.1595800,002.7085.4168.125Prefill (tok/s)Generation (tok/s)NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition - 1.799,4 tok/s Generation, 6.587 tok/s Prefill, TTFT 3.899 ms (9 Laufe)NVIDIA RTX PRO 6000 B...NVIDIA RTX PRO 6000 Blackwell Workstation Edition - 1.607,5 tok/s Generation, 6.349 tok/s Prefill, TTFT 3.508 ms (3 Laufe)NVIDIA RTX PRO 6000 B...NVIDIA GeForce RTX 5090 - 1.556,7 tok/s Generation, 5.085 tok/s Prefill, TTFT 3.918 ms (3 Laufe)NVIDIA GeForce RTX 50...NVIDIA GeForce RTX 3090 Ti - 920,2 tok/s Generation, 3.340 tok/s Prefill, TTFT 6.459 ms (3 Laufe)NVIDIA GeForce RTX 30...NVIDIA GeForce RTX 5070 Ti - 309,0 tok/s Generation, 972 tok/s Prefill, TTFT 20.166 ms (3 Laufe)NVIDIA GeForce RTX 50...AMD Radeon AI PRO R9700 - 104,8 tok/s Generation, 240 tok/s Prefill, TTFT 69.177 ms (9 Laufe) | DIESER LAUF★ AMD Radeon AI PRO R97...
NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition 1.799,4 tok/sNVIDIA RTX PRO 6000 Blackwell Workstation Edition 1.607,5 tok/sNVIDIA GeForce RTX 5090 1.556,7 tok/sNVIDIA GeForce RTX 3090 Ti 920,2 tok/sNVIDIA GeForce RTX 5070 Ti 309,0 tok/s★ AMD Radeon AI PRO R9700 104,8 tok/s this run

CPUby processor

2.3181.7391.1595800,002.7085.4168.125Prefill (tok/s)Generation (tok/s)AMD Ryzen Threadripper PRO 9965WX 24-Cores - 1.799,4 tok/s Generation, 6.587 tok/s Prefill, TTFT 3.899 ms (9 Laufe)AMD Ryzen Threadrippe...AMD Ryzen 9 9950X 16-Core Processor - 1.607,5 tok/s Generation, 6.349 tok/s Prefill, TTFT 3.508 ms (3 Laufe)AMD Ryzen 9 9950X 16-...AMD Ryzen 7 5800X3D 8-Core Processor - 1.556,7 tok/s Generation, 5.085 tok/s Prefill, TTFT 3.918 ms (3 Laufe)AMD Ryzen 7 5800X3D 8...AMD Ryzen 9 8945HX with Radeon Graphics - 920,2 tok/s Generation, 3.340 tok/s Prefill, TTFT 6.459 ms (3 Laufe)AMD Ryzen 9 8945HX wi...AMD Ryzen Threadripper PRO 5975WX 32-Cores - 309,0 tok/s Generation, 972 tok/s Prefill, TTFT 20.166 ms (3 Laufe)AMD Ryzen Threadrippe...AMD Ryzen Threadripper PRO 7955WX 16-Cores - 104,8 tok/s Generation, 240 tok/s Prefill, TTFT 69.177 ms (9 Laufe) | DIESER LAUF★ AMD Ryzen Threadrippe...
AMD Ryzen Threadripper PRO 9965WX 24-Cores 1.799,4 tok/sAMD Ryzen 9 9950X 16-Core Processor 1.607,5 tok/sAMD Ryzen 7 5800X3D 8-Core Processor 1.556,7 tok/sAMD Ryzen 9 8945HX with Radeon Graphics 920,2 tok/sAMD Ryzen Threadripper PRO 5975WX 32-Cores 309,0 tok/s★ AMD Ryzen Threadripper PRO 7955WX 16-Cores 104,8 tok/s this run

MBby mainboard

2.2771.7081.1395690,002.5665.1327.698Prefill (tok/s)Generation (tok/s)ASUSTeK COMPUTER INC. ProArt X870E-CREATOR WIFI - 1.607,5 tok/s Generation, 6.349 tok/s Prefill, TTFT 3.508 ms (3 Laufe)ASUSTeK COMPUTER INC....ASUSTeK COMPUTER INC. ROG STRIX B550-A GAMING - 1.556,7 tok/s Generation, 5.085 tok/s Prefill, TTFT 3.918 ms (3 Laufe)ASUSTeK COMPUTER INC....Meigao Innovation Technology (Shen Zhen) Co., Ltd DRFXL (MotherBoard Series) - 920,2 tok/s Generation, 3.340 tok/s Prefill, TTFT 6.459 ms (3 Laufe)Meigao Innovation Tec...ASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI - 309,0 tok/s Generation, 972 tok/s Prefill, TTFT 20.166 ms (3 Laufe)ASUSTeK COMPUTER INC....ASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE - 1.799,4 tok/s Generation, 3.413 tok/s Prefill, TTFT 36.538 ms (18 Laufe) | DIESER LAUF★ ASUSTeK COMPUTER INC....
★ ASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE 1.799,4 tok/s this runASUSTeK COMPUTER INC. ProArt X870E-CREATOR WIFI 1.607,5 tok/sASUSTeK COMPUTER INC. ROG STRIX B550-A GAMING 1.556,7 tok/sMeigao Innovation Technology (Shen Zhen) Co., Ltd DRFXL (MotherBoard Series) 920,2 tok/sASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI 309,0 tok/s

ENGby engine

1.9791.8891.7991.7091.6193.4053.5503.6953.840Prefill (tok/s)Generation (tok/s)llama.cpp - 1.799,4 tok/s Generation, 3.623 tok/s Prefill, TTFT 25.328 ms (30 Laufe) | DIESER LAUF★ llama.cpp
★ llama.cpp 1.799,4 tok/s this run

DRVby driver

1.9791.8891.7991.7091.6193.4053.5503.6953.840Prefill (tok/s)Generation (tok/s)unbekannt - 1.799,4 tok/s Generation, 3.623 tok/s Prefill, TTFT 25.328 ms (30 Laufe)unbekannt
unbekannt 1.799,4 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 0.49
Token / kWh609.48K
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)3.96B
☁️ 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

Ornith-1.0-35BAMD Radeon AI PRO R9700Ornith-1.0-35BNVIDIA RTX PRO 6000 Blackwell Workstation EditionOrnith-1.0-35BNVIDIA GeForce RTX 5090Ornith-1.0-35B3x 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.