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

Devstral-Small-2507

Performance benchmark · measured on 27.07.2026 17:54

Benchmark-IDrun-20260727-181606-e355ea
Timebench 3 - Kombi (Prefill + Generation)Dense24BRuntime: llama.cppQuantisierung: Q4_K_M
Generation308,08tok/s
Prefill16.739,26tok/s
Time to First Token3.490,50ms
Total duration41,27s
Concurrency5parallel
Ranking in the field
45of 155 systems

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

This run is better than 71 % of all comparable systems.
Generation 308,1 tok/s
+28 % vs Ø 241,1
Prefill 16.739,3 tok/s
+214 % vs Ø 5.336,0
Time to First Token 3.491 ms
-87 % vs Ø 26.930
Distribution in the field0 – 1.349 tok/s
Ø 240 Median Dieser Lauf

Wie schlägt sich dieser Benchmark mit anderen Modellen?

How does this benchmark compare on other GPUs?

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

Hardware

GPU: 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 96 GB VRAM
CPU: AMD Ryzen Threadripper PRO 9965WX 24-Cores
RAM: 125 GB
Mainboard: ASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE

Setup

Runtime: llama.cpp
Quantization: Q4_K_M
Model: Devstral-Small-2507

Configuration

benchmark-konfiguration — run-20260727-181606-e355ea
# LLM-Benchmark Konfiguration # Modell : Devstral-Small-2507 # Engine : llama.cpp # Run-ID : run-20260727-181606-e355ea # GPU : 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition # CPU : AMD Ryzen Threadripper PRO 9965WX 24-Cores # RAM : 125 GB bench@llm-benchmark:~$ /home/godcore/llama.cpp/build/bin/llama-server \ -m /home/godcore/.cache/huggingface/hub/models--mistralai--Devstral-Small-2507_gguf/snapshots/ee2f0c00c5c86862f471fbf533268cf01b80d4a6/Devstral-Small-2507-Q4_K_M.gguf \ --alias Devstral-Small-2507 \ --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.Devstral-Small-2507
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--mistralai--Devstral-Small-2507_gguf/snapshots/ee2f0c00c5c86862f471fbf533268cf01b80d4a6/Devstral-Small-2507-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

Model comparison

Devstral-Small-2507 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

7235423621810,005.46510.93116.396Prefill (tok/s)Generation (tok/s)NVIDIA GeForce RTX 3090 Ti - 317,4 tok/s Generation, 4.619 tok/s Prefill, TTFT 10.835 ms (3 Laufe)NVIDIA GeForce RTX 30...NVIDIA GeForce RTX 5070 Ti - 103,9 tok/s Generation, 2.732 tok/s Prefill, TTFT 42.889 ms (2 Laufe)NVIDIA GeForce RTX 50...NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition - 572,2 tok/s Generation, 13.619 tok/s Prefill, TTFT 5.186 ms (3 Laufe) | DIESER LAUF★ NVIDIA RTX PRO 6000 B...
★ NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition 572,2 tok/s this runNVIDIA GeForce RTX 3090 Ti 317,4 tok/sNVIDIA GeForce RTX 5070 Ti 103,9 tok/s

CPUby processor

7235423621810,005.46510.93116.396Prefill (tok/s)Generation (tok/s)AMD Ryzen 9 8945HX with Radeon Graphics - 317,4 tok/s Generation, 4.619 tok/s Prefill, TTFT 10.835 ms (3 Laufe)AMD Ryzen 9 8945HX wi...AMD Ryzen Threadripper PRO 5975WX 32-Cores - 103,9 tok/s Generation, 2.732 tok/s Prefill, TTFT 42.889 ms (2 Laufe)AMD Ryzen Threadrippe...AMD Ryzen Threadripper PRO 9965WX 24-Cores - 572,2 tok/s Generation, 13.619 tok/s Prefill, TTFT 5.186 ms (3 Laufe) | DIESER LAUF★ AMD Ryzen Threadrippe...
★ AMD Ryzen Threadripper PRO 9965WX 24-Cores 572,2 tok/s this runAMD Ryzen 9 8945HX with Radeon Graphics 317,4 tok/sAMD Ryzen Threadripper PRO 5975WX 32-Cores 103,9 tok/s

MBby mainboard

7235423621810,005.46510.93116.396Prefill (tok/s)Generation (tok/s)Meigao Innovation Technology (Shen Zhen) Co., Ltd DRFXL (MotherBoard Series) - 317,4 tok/s Generation, 4.619 tok/s Prefill, TTFT 10.835 ms (3 Laufe)Meigao Innovation Tec...ASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI - 103,9 tok/s Generation, 2.732 tok/s Prefill, TTFT 42.889 ms (2 Laufe)ASUSTeK COMPUTER INC....ASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE - 572,2 tok/s Generation, 13.619 tok/s Prefill, TTFT 5.186 ms (3 Laufe) | DIESER LAUF★ ASUSTeK COMPUTER INC....
★ ASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE 572,2 tok/s this runMeigao Innovation Technology (Shen Zhen) Co., Ltd DRFXL (MotherBoard Series) 317,4 tok/sASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI 103,9 tok/s

ENGby engine

6296015725445157.0717.3727.6737.974Prefill (tok/s)Generation (tok/s)llama.cpp - 572,2 tok/s Generation, 7.522 tok/s Prefill, TTFT 16.730 ms (8 Laufe) | DIESER LAUF★ llama.cpp
★ llama.cpp 572,2 tok/s this run

DRVby driver

6296015725445157.0717.3727.6737.974Prefill (tok/s)Generation (tok/s)unbekannt - 572,2 tok/s Generation, 7.522 tok/s Prefill, TTFT 16.730 ms (8 Laufe)unbekannt
unbekannt 572,2 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 165 W
⚡ TDP 983 W
☁️ External LLM (API)
Electricity0.30 EUR/kWh
Avg power (incl. idle)983 W estimated (TDP)GPU 900 + CPU 57 + Board 25 W full load
Avg cost / hourEUR 0.29
Electricity / 1M tokensEUR 0.27
Token / kWh1.13M
Acquisition (system)EUR 45,748 full priceGPU EUR 39,000 · CPU EUR 3,499 · Board EUR 1,299 · RAM EUR 1,750 · PSU EUR 200
Electricity (2 years)
TCO (2 years)EUR 50,912
Output tokens (2 years)19.43B
☁️ 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 (165 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

Devstral-Small-25073x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionDevstral-Small-2507NVIDIA GeForce RTX 3090 TiDevstral-Small-2507NVIDIA 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.