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

Ling-3.0-flash

Performance benchmark · measured on 20.08.2026 06:44

Benchmark-IDrun-20260820-071726-fac92e
Timebench 3 - Kombi (Prefill + Generation)MoE127.5BRuntime: llama.cppQuantisierung: Q4_K_M
ℹ
For context: Diese Plattform nutzt Unified Memory – die "VRAM" ist gemeinsamer System-RAM (APU/Superchip); das Modell teilt sich den Speicher mit dem System.
Generation90,05tok/s
Prefill1.599,60tok/s
Time to First Token43.239,00ms
Total duration372,71s
Concurrency10parallel
Ranking in the field
1137of 1427 systems

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

This run is better than 20 % of all comparable systems.
Generation 90,1 tok/s
-79 % vs Ø 425,3
Prefill 1.599,6 tok/s
-76 % vs Ø 6.659,3
Time to First Token 43.239 ms
-6 % vs Ø 45.848
Distribution in the field0 – 2.491 tok/s
Ø 425 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-3.5-Lightning-30B-A3BNVIDIA GeForce RTX 3090 Ti · run-20261021-093815-dce558
2.263,7 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× 3x 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× 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-194136-0aed86
1.937,7 tok/s
Nemotron-3-Nano-4B3× 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-194136-8db0fa
1.911,6 tok/s
Nemotron-3-Nano-4B3× 3x 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× 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-034052-26063b
1.835,7 tok/s
Nemotron-3-Nano-4B3× 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · run-20260728-194136-a77046
1.830,3 tok/s
Ling-3.0-flash this runNVIDIA GB10 (DGX Spark) · run-20260820-071726-fac92e
90,1 tok/s

How does this benchmark compare on other GPUs?

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

Hardware

GPU: NVIDIA GB10 · 128 GB VRAM
CPU: NVIDIA Grace
RAM: 120 GB
Mainboard: ASUSTeK COMPUTER INC. GX10

Setup

Runtime: llama.cpp
Quantization: Q4_K_M
Driver: NVIDIA 590.48.01 / CUDA 13.1
Model: Ling-3.0-flash

Configuration

benchmark-konfiguration — run-20260820-071726-fac92e
# LLM-Benchmark Konfiguration # Modell : Ling-3.0-flash # Engine : llama.cpp # Run-ID : run-20260820-071726-fac92e # GPU : NVIDIA GB10 # CPU : NVIDIA Grace # RAM : 120 GB bench@llm-benchmark:~$ /home/godcore/llama.cpp/build/bin/llama-server \ -m /home/godcore/ling_gguf/Ling-3.0-flash-Q4_K_M.gguf \ --alias Ling-3.0-flash \ -ngl 999 \ -fa on \ -c 49152 \ -np 12 \ -b 8192 \ -ub 4096 \ --jinja \ --host 0.0.0.0 \ --port 8000
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.Ling-3.0-flash
Kontextlaenge?Maximale Anzahl Tokens (Eingabe + erzeugte Ausgabe zusammen), die das Modell pro Anfrage verarbeiten kann.49152
Modellpfad?Pfad zur GGUF-Modelldatei, die geladen und ausgeliefert wird./home/godcore/ling_gguf/Ling-3.0-flash-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.49152
np12
Batch-Groesse?Logische Batchgroesse fuer die Prompt-Verarbeitung (Anzahl Token pro Prefill-Durchlauf).8192
Micro-Batch?Physische Micro-Batchgroesse. Begrenzt, wie viele Token gleichzeitig durch die Compute-Pipeline laufen.4096
execution_typelocal

All benchmarks of this model To leaderboard

Anzeige
Model comparison

Ling-3.0-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

47836926115243,101.3262.6523.978Prefill (tok/s)Generation (tok/s)NVIDIA RTX A6000 - 388,1 tok/s Generation, 3.235 tok/s Prefill, TTFT 7.234 ms (15 Laufe)NVIDIA RTX A6000AMD Radeon PRO W7900 Dual Slot - 165,7 tok/s Generation, 1.163 tok/s Prefill, TTFT 7.103 ms (27 Laufe)AMD Radeon PRO W7900 ...AMD Radeon AI PRO R9700 - 160,2 tok/s Generation, 1.407 tok/s Prefill, TTFT 6.543 ms (23 Laufe)AMD Radeon AI PRO R97...AMD Radeon RX 7900 XTX - 133,0 tok/s Generation, 184 tok/s Prefill, TTFT 122.114 ms (7 Laufe)AMD Radeon RX 7900 XTXNVIDIA GB10 (DGX Spark) - 253,7 tok/s Generation, 1.278 tok/s Prefill, TTFT 22.647 ms (15 Laufe) | DIESER LAUF★ NVIDIA GB10 (DGX Spar...
NVIDIA RTX A6000 388,1 tok/s★ NVIDIA GB10 (DGX Spark) 253,7 tok/s this runAMD Radeon PRO W7900 Dual Slot 165,7 tok/sAMD Radeon AI PRO R9700 160,2 tok/sAMD Radeon RX 7900 XTX 133,0 tok/s

CPUby processor

5013762511250,001.3302.6613.991Prefill (tok/s)Generation (tok/s)AMD EPYC 7203P 8-Core Processor - 388,1 tok/s Generation, 3.235 tok/s Prefill, TTFT 7.234 ms (15 Laufe)AMD EPYC 7203P 8-Core...AMD Ryzen Threadripper PRO 5975WX 32-Cores - 165,7 tok/s Generation, 1.163 tok/s Prefill, TTFT 7.103 ms (27 Laufe)AMD Ryzen Threadrippe...AMD Ryzen Threadripper PRO 7955WX 16-Cores - 160,2 tok/s Generation, 1.407 tok/s Prefill, TTFT 6.543 ms (23 Laufe)AMD Ryzen Threadrippe...AMD Ryzen Threadripper PRO 3955WX 16-Cores - 133,0 tok/s Generation, 240 tok/s Prefill, TTFT 131.640 ms (4 Laufe)AMD Ryzen Threadrippe...Intel(R) Xeon(R) CPU E5-4657L v2 @ 2.40GHz - 14,9 tok/s Generation, 109 tok/s Prefill, TTFT 109.414 ms (3 Laufe)Intel(R) Xeon(R) CPU ...NVIDIA Grace - 253,7 tok/s Generation, 1.278 tok/s Prefill, TTFT 22.647 ms (15 Laufe) | DIESER LAUF★ NVIDIA Grace
AMD EPYC 7203P 8-Core Processor 388,1 tok/s★ NVIDIA Grace 253,7 tok/s this runAMD Ryzen Threadripper PRO 5975WX 32-Cores 165,7 tok/sAMD Ryzen Threadripper PRO 7955WX 16-Cores 160,2 tok/sAMD Ryzen Threadripper PRO 3955WX 16-Cores 133,0 tok/sIntel(R) Xeon(R) CPU E5-4657L v2 @ 2.40GHz 14,9 tok/s

MBby mainboard

5013762511250,001.8743.7485.622Prefill (tok/s)Generation (tok/s)ASRockRack ROMED8-2T/BCM - 388,1 tok/s Generation, 4.550 tok/s Prefill, TTFT 4.923 ms (9 Laufe)ASRockRack ROMED8-2T/...ASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI - 194,7 tok/s Generation, 1.079 tok/s Prefill, TTFT 21.150 ms (37 Laufe)ASUSTeK COMPUTER INC....ASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE - 160,2 tok/s Generation, 1.407 tok/s Prefill, TTFT 6.543 ms (23 Laufe)ASUSTeK COMPUTER INC....Dell Inc. PowerEdge R820 - 14,9 tok/s Generation, 109 tok/s Prefill, TTFT 109.414 ms (3 Laufe)Dell Inc. PowerEdge R...ASUSTeK COMPUTER INC. GX10 - 253,7 tok/s Generation, 1.278 tok/s Prefill, TTFT 22.647 ms (15 Laufe) | DIESER LAUF★ ASUSTeK COMPUTER INC....
ASRockRack ROMED8-2T/BCM 388,1 tok/s★ ASUSTeK COMPUTER INC. GX10 253,7 tok/s this runASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI 194,7 tok/sASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE 160,2 tok/sDell Inc. PowerEdge R820 14,9 tok/s

ENGby engine

45438732125418804.0978.19312.290Prefill (tok/s)Generation (tok/s)vLLM - 388,1 tok/s Generation, 10.088 tok/s Prefill, TTFT 1.544 ms (3 Laufe)vLLMllama.cpp - 253,7 tok/s Generation, 1.220 tok/s Prefill, TTFT 19.532 ms (84 Laufe) | DIESER LAUF★ llama.cpp
vLLM 388,1 tok/s★ llama.cpp 253,7 tok/s this run

DRVby driver

4543873212541881.1291.3281.5271.726Prefill (tok/s)Generation (tok/s)unbekannt - 388,1 tok/s Generation, 1.577 tok/s Prefill, TTFT 18.133 ms (72 Laufe)unbekanntNVIDIA 590.48.01 / CUDA 13.1 - 253,7 tok/s Generation, 1.278 tok/s Prefill, TTFT 22.647 ms (15 Laufe) | DIESER LAUF★ NVIDIA 590.48.01 / CU...
unbekannt 388,1 tok/s★ NVIDIA 590.48.01 / CUDA 13.1 253,7 tok/s this run
💰 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 15 W
⚡ TDP 140 W
☁️ External LLM (API)
Electricity0.30 EUR/kWh
Avg power (incl. idle)140 W estimated (TDP)GPU 140 W full load
Avg cost / hourEUR 0.042
Electricity / 1M tokensEUR 0.13
Token / kWh2.32M
Acquisition (system)EUR 5,800 partial priceGPU EUR 4,000 · RAM EUR 1,680 · PSU EUR 120
Electricity (2 years)–
TCO (2 years)EUR 6,536
Output tokens (2 years)5.68B
☁️ 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 (15 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

Ling-3.0-flashNVIDIA GB10 (DGX Spark)Ling-3.0-flash2x NVIDIA RTX A6000Ling-3.0-flashNVIDIA GB10 (DGX Spark)Ling-3.0-flashNVIDIA GB10 (DGX Spark)
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.