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

Qwen3.8-Flash-Next

Performance benchmark · measured on 23.10.2026 08:30

Benchmark-IDrun-20261023-180245-c8d477
Timebench 3 - Kombi (Prefill + Generation)MoE180BRuntime: vLLMQuantisierung: NVFP4
ℹ
Public note: NVIDIA RTX PRO 6000 Blackwell Workstation Edition 96GB . Qwen3.8-Flash-Next-Uncensored NVFP4 (mazinb) . vLLM, max-model-len 131072, max-num-seqs 10, gpu-memory-utilization 0.90, Reasoning aktiv
⚠
Configuration note: Es sind 2 Grafikkarten installiert, aber nur 1 wurde fuer diesen Lauf genutzt. Mit allen 2 GPUs koennte der Durchsatz deutlich hoeher liegen.
Generation584,06tok/s
Prefill25.275,57tok/s
Time to First Token1.074,50ms
Total duration37,26s
Concurrency10parallel
Ranking in the field
362of 1464 systems

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

This run is better than 75 % of all comparable systems.
Generation 584,1 tok/s
+31 % vs Ø 447,2
Prefill 25.275,6 tok/s
+269 % vs Ø 6.843,8
Time to First Token 1.075 ms
-98 % vs Ø 44.831
Distribution in the field0 – 2.491 tok/s
Ø 447 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
gemma-4-E2B-itNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20261023-065752-48370d
2.412,3 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
Nemotron-3-Nano-4BNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20261023-065758-96fe6a
2.144,3 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
gpt-oss-20bNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20261023-065808-332d1f
1.981,1 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
Qwen3.8-Flash-Next this runNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20261023-180245-c8d477
584,1 tok/s

How does this benchmark compare on other GPUs?

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

Hardware

GPU: NVIDIA RTX PRO 6000 Blackwell Workstation Edition · 96 GB VRAM
CPU: AMD Ryzen 9 9950X 16-Core Processor
RAM: 123 GB
Mainboard: ASUSTeK COMPUTER INC. ProArt X870E-CREATOR WIFI

Setup

Runtime: vLLM
Quantization: NVFP4
Model: Qwen3.8-Flash-Next

Configuration

benchmark-konfiguration — run-20261023-180245-c8d477
# LLM-Benchmark Konfiguration # Modell : Qwen3.8-Flash-Next # Engine : vLLM # Run-ID : run-20261023-180245-c8d477 # GPU : NVIDIA RTX PRO 6000 Blackwell Workstation Edition # CPU : AMD Ryzen 9 9950X 16-Core Processor # RAM : 123 GB bench@llm-benchmark:~$ /usr/bin/python3 /usr/local/bin/vllm serve mazinb/Qwen3.8-Flash-Next-Uncensored-NVFP4 \ --served-model-name Qwen3.8-Flash-Next-Uncensored \ --host 0.0.0.0 \ --port 8000 \ --trust-remote-code \ --distributed-executor-backend mp \ --gpu-memory-utilization 0.90 \ --max-model-len 131072 \ --max-num-seqs 10 \ --enable-auto-tool-choice \ --tool-call-parser qwen3_coder \ --reasoning-parser qwen3
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.vllm
Modellalias?Der Name, unter dem das Modell ueber die API angesprochen wird. Genau dieser Wert muss im Request-Feld 'model' stehen.Qwen3.8-Flash-Next-Uncensored
Kontextlaenge?Maximale Anzahl Tokens (Eingabe + erzeugte Ausgabe zusammen), die das Modell pro Anfrage verarbeiten kann.131072
Alias?Anzeigename des Modells nach aussen (served model name), unabhaengig vom Dateinamen.Qwen3.8-Flash-Next-Uncensored
Trust-Remote-Code?Erlaubt vLLM, mit dem Modell mitgelieferten Python-Code auszufuehren. Noetig fuer Architekturen, die vLLM nicht von Haus aus kennt. Nur bei vertrauenswuerdigen Modellen aktivieren.aktiv
Executor-Backend?Verfahren fuer Multi-GPU/Multi-Node: mp = Prozesse auf einer Maschine, ray = ueber Ray (auch ueber mehrere Knoten).mp
GPU-Speicher?Anteil des GPU-Speichers (0 bis 1), den vLLM belegen darf. 0.92 = 92 %. Hoeher = mehr Platz fuer den KV-Cache (mehr/laengere parallele Anfragen), aber groesseres Risiko fuer 'Out of Memory'.0.90
Kontext?Groesse des Kontextfensters in Token. 0 = der beim Training verwendete Kontext des Modells.131072
Max-Sequenzen?Maximale Anzahl gleichzeitig bearbeiteter Anfragen (Batch-Groesse). Hoeher = mehr Durchsatz, braucht aber mehr KV-Cache-Speicher.10
Auto-Tool-Choice?Erlaubt dem Modell, Tools (Function-Calling) selbststaendig auszuwaehlen. MUSS aktiv sein, damit 'tools' im Request akzeptiert werden - sonst lehnt vLLM sie mit HTTP 400 ab.aktiv
Tool-Parser?Bestimmt, wie vLLM Tool-Aufrufe aus der Modellantwort ausliest (z.B. hermes, mistral, llama3_json). Muss zum Modell passen.qwen3_coder
Reasoning-Parser?Trennt den Denk-/Reasoning-Teil der Antwort vom eigentlichen Inhalt (z.B. bei Modellen mit <think>-Bloecken).qwen3
execution_typelocal

All benchmarks of this model To leaderboard

Anzeige
Model comparison

Qwen3.8-Flash-Next 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

2.0301.5231.0155080,005.77211.54417.316Prefill (tok/s)Generation (tok/s)NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition - 1.564,6 tok/s Generation, 6.065 tok/s Prefill, TTFT 7.635 ms (15 Laufe)NVIDIA RTX PRO 6000 B...NVIDIA RTX A6000 - 1.171,8 tok/s Generation, 963 tok/s Prefill, TTFT 32.556 ms (21 Laufe)NVIDIA RTX A6000NVIDIA GeForce RTX 3090 - 1.132,7 tok/s Generation, 968 tok/s Prefill, TTFT 51.682 ms (6 Laufe)NVIDIA GeForce RTX 30...NVIDIA GB10 (DGX Spark) - 492,8 tok/s Generation, 1.262 tok/s Prefill, TTFT 33.373 ms (12 Laufe)NVIDIA GB10 (DGX Spar...NVIDIA Tesla V100 PCIe 32GB - 196,2 tok/s Generation, 775 tok/s Prefill, TTFT 30.834 ms (8 Laufe)NVIDIA Tesla V100 PCI...AMD Radeon RX 7900 XTX - 18,7 tok/s Generation, 180 tok/s Prefill, TTFT 55.956 ms (3 Laufe)AMD Radeon RX 7900 XTXNVIDIA RTX PRO 6000 Blackwell Workstation Edition - 584,1 tok/s Generation, 13.991 tok/s Prefill, TTFT 512 ms (5 Laufe) | DIESER LAUF★ NVIDIA RTX PRO 6000 B...
NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition 1.564,6 tok/sNVIDIA RTX A6000 1.171,8 tok/sNVIDIA GeForce RTX 3090 1.132,7 tok/s★ NVIDIA RTX PRO 6000 Blackwell Workstation Edition 584,1 tok/s this runNVIDIA GB10 (DGX Spark) 492,8 tok/sNVIDIA Tesla V100 PCIe 32GB 196,2 tok/sAMD Radeon RX 7900 XTX 18,7 tok/s

CPUby processor

2.0301.5231.0155080,005.78311.56617.348Prefill (tok/s)Generation (tok/s)AMD Ryzen 9 7945HX with Radeon Graphics - 1.564,6 tok/s Generation, 2.312 tok/s Prefill, TTFT 17.248 ms (6 Laufe)AMD Ryzen 9 7945HX wi...AMD Ryzen Threadripper PRO 3945WX 12-Cores - 1.171,8 tok/s Generation, 2.569 tok/s Prefill, TTFT 15.123 ms (6 Laufe)AMD Ryzen Threadrippe...AMD Ryzen 5 5600X 6-Core Processor - 1.132,7 tok/s Generation, 5.169 tok/s Prefill, TTFT 34.723 ms (9 Laufe)AMD Ryzen 5 5600X 6-C...NVIDIA Grace - 492,8 tok/s Generation, 1.262 tok/s Prefill, TTFT 33.373 ms (12 Laufe)NVIDIA GraceAMD Ryzen Threadripper PRO 9965WX 24-Cores - 424,2 tok/s Generation, 0 tok/s Prefill, TTFT 1.438 ms (6 Laufe)AMD Ryzen Threadrippe...AMD Ryzen 3 PRO 3200GE w/ Radeon Vega Graphics - 196,2 tok/s Generation, 757 tok/s Prefill, TTFT 36.737 ms (5 Laufe)AMD Ryzen 3 PRO 3200G...AMD EPYC 7203P 8-Core Processor - 79,2 tok/s Generation, 320 tok/s Prefill, TTFT 39.530 ms (15 Laufe)AMD EPYC 7203P 8-Core...Intel(R) Xeon(R) CPU E5-4657L v2 @ 2.40GHz - 63,2 tok/s Generation, 803 tok/s Prefill, TTFT 20.996 ms (3 Laufe)Intel(R) Xeon(R) CPU ...AMD Ryzen Threadripper PRO 3955WX 16-Cores - 18,7 tok/s Generation, 180 tok/s Prefill, TTFT 55.956 ms (3 Laufe)AMD Ryzen Threadrippe...AMD Ryzen 9 9950X 16-Core Processor - 584,1 tok/s Generation, 13.991 tok/s Prefill, TTFT 512 ms (5 Laufe) | DIESER LAUF★ AMD Ryzen 9 9950X 16-...
AMD Ryzen 9 7945HX with Radeon Graphics 1.564,6 tok/sAMD Ryzen Threadripper PRO 3945WX 12-Cores 1.171,8 tok/sAMD Ryzen 5 5600X 6-Core Processor 1.132,7 tok/s★ AMD Ryzen 9 9950X 16-Core Processor 584,1 tok/s this runNVIDIA Grace 492,8 tok/sAMD Ryzen Threadripper PRO 9965WX 24-Cores 424,2 tok/sAMD Ryzen 3 PRO 3200GE w/ Radeon Vega Graphics 196,2 tok/sAMD EPYC 7203P 8-Core Processor 79,2 tok/sIntel(R) Xeon(R) CPU E5-4657L v2 @ 2.40GHz 63,2 tok/sAMD Ryzen Threadripper PRO 3955WX 16-Cores 18,7 tok/s

MBby mainboard

2.0271.5201.0135070,005.78311.56617.348Prefill (tok/s)Generation (tok/s)Shenzhen Meigao Electronic Equipment Co.,Ltd DRFXI (MotherBoard Series) - 1.564,6 tok/s Generation, 2.312 tok/s Prefill, TTFT 17.248 ms (6 Laufe)Shenzhen Meigao Elect...unbekannt - 1.171,8 tok/s Generation, 2.569 tok/s Prefill, TTFT 15.123 ms (6 Laufe)unbekanntASUSTeK COMPUTER INC. ROG STRIX B550-A GAMING - 1.132,7 tok/s Generation, 5.169 tok/s Prefill, TTFT 34.723 ms (9 Laufe)ASUSTeK COMPUTER INC....ASUSTeK COMPUTER INC. GX10 - 492,8 tok/s Generation, 1.262 tok/s Prefill, TTFT 33.373 ms (12 Laufe)ASUSTeK COMPUTER INC....ASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE - 424,2 tok/s Generation, 0 tok/s Prefill, TTFT 1.438 ms (6 Laufe)ASUSTeK COMPUTER INC....ASUSTeK COMPUTER INC. ROG STRIX X570-F GAMING - 196,2 tok/s Generation, 757 tok/s Prefill, TTFT 36.737 ms (5 Laufe)ASUSTeK COMPUTER INC....ASRockRack ROMED8-2T/BCM - 79,2 tok/s Generation, 423 tok/s Prefill, TTFT 24.603 ms (9 Laufe)ASRockRack ROMED8-2T/...Dell Inc. PowerEdge R820 - 63,2 tok/s Generation, 803 tok/s Prefill, TTFT 20.996 ms (3 Laufe)Dell Inc. PowerEdge R...ASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI - 35,4 tok/s Generation, 170 tok/s Prefill, TTFT 59.933 ms (9 Laufe)ASUSTeK COMPUTER INC....ASUSTeK COMPUTER INC. ProArt X870E-CREATOR WIFI - 584,1 tok/s Generation, 13.991 tok/s Prefill, TTFT 512 ms (5 Laufe) | DIESER LAUF★ ASUSTeK COMPUTER INC....
Shenzhen Meigao Electronic Equipment Co.,Ltd DRFXI (MotherBoard Series) 1.564,6 tok/sunbekannt 1.171,8 tok/sASUSTeK COMPUTER INC. ROG STRIX B550-A GAMING 1.132,7 tok/s★ ASUSTeK COMPUTER INC. ProArt X870E-CREATOR WIFI 584,1 tok/s this runASUSTeK COMPUTER INC. GX10 492,8 tok/sASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE 424,2 tok/sASUSTeK COMPUTER INC. ROG STRIX X570-F GAMING 196,2 tok/sASRockRack ROMED8-2T/BCM 79,2 tok/sDell Inc. PowerEdge R820 63,2 tok/sASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI 35,4 tok/s

ENGby engine

1.9171.4961.07465323104.2738.54612.819Prefill (tok/s)Generation (tok/s)strata - 1.564,6 tok/s Generation, 1.064 tok/s Prefill, TTFT 44.996 ms (20 Laufe)stratallama.cpp - 1.555,3 tok/s Generation, 627 tok/s Prefill, TTFT 35.323 ms (24 Laufe)llama.cppunbekannt - 1.171,8 tok/s Generation, 1.981 tok/s Prefill, TTFT 17.080 ms (9 Laufe)unbekanntvLLM - 584,1 tok/s Generation, 10.429 tok/s Prefill, TTFT 1.500 ms (17 Laufe) | DIESER LAUF★ vLLM
strata 1.564,6 tok/sllama.cpp 1.555,3 tok/sunbekannt 1.171,8 tok/s★ vLLM 584,1 tok/s this run

DRVby driver

1.9351.4821.0295751228151.6292.4433.257Prefill (tok/s)Generation (tok/s)unbekannt - 1.564,6 tok/s Generation, 2.810 tok/s Prefill, TTFT 26.753 ms (58 Laufe)unbekanntNVIDIA 590.48.01 / CUDA 13.1 - 492,8 tok/s Generation, 1.262 tok/s Prefill, TTFT 33.373 ms (12 Laufe)NVIDIA 590.48.01 / CU...
unbekannt 1.564,6 tok/sNVIDIA 590.48.01 / CUDA 13.1 492,8 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 70 W
⚡ TDP 644 W
☁️ External LLM (API)
Electricity0.30 EUR/kWh
Avg power (incl. idle)644 W estimated (TDP)GPU 600 + CPU 29 + Board 15 W full load
Avg cost / hourEUR 0.19
Electricity / 1M tokensEUR 0.092
Token / kWh3.27M
Acquisition (system)EUR 15,558 full priceGPU EUR 13,000 · CPU EUR 649 · Board EUR 499 · RAM EUR 1,230 · PSU EUR 180
Electricity (2 years)–
TCO (2 years)EUR 18,942
Output tokens (2 years)36.84B
☁️ 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 (70 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

Qwen3.8-Flash-NextNVIDIA RTX PRO 6000 Blackwell Workstation EditionQwen3.8-Flash-NextNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionQwen3.8-Flash-NextNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionQwen3.8-Flash-NextNVIDIA RTX A6000
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.