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

Mistral-Medium-3.5-128B

Performance benchmark · measured on 29.07.2026 06:35

Benchmark-IDrun-20260729-105332-853af7
Timebench 3 - Kombi (Prefill + Generation)Dense128BRuntime: llama.cppQuantisierung: Q4_K_M
Generation73,58tok/s
Prefill1.876,84tok/s
Time to First Token22.217,50ms
Total duration271,49s
Concurrency5parallel
Ranking in the field
323of 432 systems

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

This run is better than 25 % of all comparable systems.
Generation 73,6 tok/s
-82 % vs Ø 402,0
Prefill 1.876,8 tok/s
-67 % vs Ø 5.763,2
Time to First Token 22.218 ms
0 % vs Ø 22.328
Distribution in the field0 – 1.349 tok/s
Ø 401 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 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
Nemotron-Cascade-2-30B-A3BNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260729-032120-4b8514
1.007,3 tok/s
Mistral-Medium-3.5-128B this runNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260729-105332-853af7
73,6 tok/s

How does this benchmark compare on other GPUs?

Same model on different hardware · 5× 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: 92 GB
Mainboard: ASUSTeK COMPUTER INC. ProArt X870E-CREATOR WIFI

Setup

Runtime: llama.cpp
Quantization: Q4_K_M
Model: Mistral-Medium-3.5-128B

Configuration

benchmark-konfiguration — run-20260729-105332-853af7
# LLM-Benchmark Konfiguration # Modell : Mistral-Medium-3.5-128B # Engine : llama.cpp # Run-ID : run-20260729-105332-853af7 # GPU : NVIDIA RTX PRO 6000 Blackwell Workstation Edition # CPU : AMD Ryzen 9 9950X 16-Core Processor # RAM : 92 GB bench@llm-benchmark:~$ /home/godcore/llama.cpp/build/bin/llama-server \ -m /home/godcore/.cache/huggingface/hub/models--unsloth--Mistral-Medium-3.5-128B-GGUF/snapshots/c8f5b1477e1b22cd2d819157d450f001f7047298/Q4_K_M/Mistral-Medium-3.5-128B-Q4_K_M-00001-of-00003.gguf \ --alias Mistral-Medium-3.5-128B \ --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.Mistral-Medium-3.5-128B
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--Mistral-Medium-3.5-128B-GGUF/snapshots/c8f5b1477e1b22cd2d819157d450f001f7047298/Q4_K_M/Mistral-Medium-3.5-128B-Q4_K_M-00001-of-00003.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

Mistral-Medium-3.5-128B 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

18313791,645,80,006951.3902.085Prefill (tok/s)Generation (tok/s)AMD Radeon AI PRO R9700 - 21,2 tok/s Generation, 1.023 tok/s Prefill, TTFT 19.761 ms (2 Laufe)AMD Radeon AI PRO R97...NVIDIA GeForce RTX 5070 Ti - 4,2 tok/s Generation, 131 tok/s Prefill, TTFT 53.481 ms (3 Laufe)NVIDIA GeForce RTX 50...NVIDIA RTX PRO 6000 Blackwell Workstation Edition - 141,6 tok/s Generation, 1.700 tok/s Prefill, TTFT 31.951 ms (3 Laufe) | DIESER LAUF★ NVIDIA RTX PRO 6000 B...
★ NVIDIA RTX PRO 6000 Blackwell Workstation Edition 141,6 tok/s this runAMD Radeon AI PRO R9700 21,2 tok/sNVIDIA GeForce RTX 5070 Ti 4,2 tok/s

CPUby processor

18313791,645,80,006951.3902.085Prefill (tok/s)Generation (tok/s)AMD Ryzen Threadripper PRO 7955WX 16-Cores - 21,2 tok/s Generation, 1.023 tok/s Prefill, TTFT 19.761 ms (2 Laufe)AMD Ryzen Threadrippe...AMD Ryzen Threadripper PRO 5975WX 32-Cores - 4,2 tok/s Generation, 131 tok/s Prefill, TTFT 53.481 ms (3 Laufe)AMD Ryzen Threadrippe...AMD Ryzen 9 9950X 16-Core Processor - 141,6 tok/s Generation, 1.700 tok/s Prefill, TTFT 31.951 ms (3 Laufe) | DIESER LAUF★ AMD Ryzen 9 9950X 16-...
★ AMD Ryzen 9 9950X 16-Core Processor 141,6 tok/s this runAMD Ryzen Threadripper PRO 7955WX 16-Cores 21,2 tok/sAMD Ryzen Threadripper PRO 5975WX 32-Cores 4,2 tok/s

MBby mainboard

18313791,645,80,006951.3902.085Prefill (tok/s)Generation (tok/s)ASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE - 21,2 tok/s Generation, 1.023 tok/s Prefill, TTFT 19.761 ms (2 Laufe)ASUSTeK COMPUTER INC....ASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI - 4,2 tok/s Generation, 131 tok/s Prefill, TTFT 53.481 ms (3 Laufe)ASUSTeK COMPUTER INC....ASUSTeK COMPUTER INC. ProArt X870E-CREATOR WIFI - 141,6 tok/s Generation, 1.700 tok/s Prefill, TTFT 31.951 ms (3 Laufe) | DIESER LAUF★ ASUSTeK COMPUTER INC....
★ ASUSTeK COMPUTER INC. ProArt X870E-CREATOR WIFI 141,6 tok/s this runASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE 21,2 tok/sASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI 4,2 tok/s

ENGby engine

156149142134127886924961999Prefill (tok/s)Generation (tok/s)llama.cpp - 141,6 tok/s Generation, 943 tok/s Prefill, TTFT 36.977 ms (8 Laufe) | DIESER LAUF★ llama.cpp
★ llama.cpp 141,6 tok/s this run

DRVby driver

156149142134127886924961999Prefill (tok/s)Generation (tok/s)unbekannt - 141,6 tok/s Generation, 943 tok/s Prefill, TTFT 36.977 ms (8 Laufe)unbekannt
unbekannt 141,6 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 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 0.73
Token / kWh411.48K
Acquisition (system)EUR 15,248 full priceGPU EUR 13,000 · CPU EUR 649 · Board EUR 499 · RAM EUR 920 · PSU EUR 180
Electricity (2 years)
TCO (2 years)EUR 18,632
Output tokens (2 years)4.64B
☁️ 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

Mistral-Medium-3.5-128BNVIDIA RTX PRO 6000 Blackwell Workstation EditionMistral-Medium-3.5-128B3x AMD Radeon AI PRO R9700Mistral-Medium-3.5-128BNVIDIA 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.