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

Devstral-Small-2-24B-Instruct-2512

Performance benchmark · measured on 28.07.2026 20:05

Benchmark-IDrun-20260729-032117-d0b46e
Timebench 3 - Kombi (Prefill + Generation)Dense24BRuntime: llama.cppQuantisierung: Q4_K_M
Generation358,30tok/s
Prefill8.068,86tok/s
Time to First Token4.298,00ms
Total duration52,65s
Concurrency5parallel
Ranking in the field
41of 61 systems

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

This run is better than 33 % of all comparable systems.
Generation 358,3 tok/s
-38 % vs Ø 578,6
Prefill 8.068,9 tok/s
-15 % vs Ø 9.517,0
Time to First Token 4.298 ms
-69 % vs Ø 13.873
Distribution in the field0 – 1.349 tok/s
Ø 579 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
NVIDIA-Nemotron-3-Nano-4BNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260727-181607-30d864
1.188,7 tok/s
gpt-oss-20bNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260728-140954-9e81f9
1.136,3 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
Nemotron-3-Nano-Omni-30B-A3B-ReasoningNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260729-032119-1ddf01
1.007,0 tok/s
Laguna-XS-2.1NVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260728-140955-17e7b6
975,1 tok/s
Laguna-XS-2.1NVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260728-140955-d94e39
967,9 tok/s
Laguna-XS-2.1NVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260728-140955-8dd350
966,1 tok/s
Qwen3-Coder-30B-A3B-InstructNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260729-032124-ebc781
938,2 tok/s
Mamba-Codestral-7B-v0.1NVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260728-194136-f3211a
927,1 tok/s
Qwen3-30B-A3B-Thinking-2507NVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260729-032123-277335
926,2 tok/s
gemma-4-E4B-itNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260727-181607-bf4219
914,3 tok/s
Qwen3-30B-A3B-Instruct-2507NVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260729-032122-601ed1
910,8 tok/s
Devstral-Small-2-24B-Instruct-2512 this runNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260729-032117-d0b46e
358,3 tok/s

How does this benchmark compare on other GPUs?

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

Configuration

benchmark-konfiguration — run-20260729-032117-d0b46e
# LLM-Benchmark Konfiguration # Modell : Devstral-Small-2-24B-Instruct-2512 # Engine : llama.cpp # Run-ID : run-20260729-032117-d0b46e # 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--Devstral-Small-2-24B-Instruct-2512-GGUF/snapshots/6e458b8add42681bfd023de5eab93637694aaf82/Devstral-Small-2-24B-Instruct-2512-Q4_K_M.gguf \ --alias Devstral-Small-2-24B-Instruct-2512 \ --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-2-24B-Instruct-2512
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--Devstral-Small-2-24B-Instruct-2512-GGUF/snapshots/6e458b8add42681bfd023de5eab93637694aaf82/Devstral-Small-2-24B-Instruct-2512-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
Model comparison

Devstral-Small-2-24B-Instruct-2512 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

9226914612300,003.2386.4779.715Prefill (tok/s)Generation (tok/s)NVIDIA GeForce RTX 3090 Ti - 331,6 tok/s Generation, 3.668 tok/s Prefill, TTFT 14.010 ms (3 Laufe)NVIDIA GeForce RTX 30...NVIDIA GeForce RTX 5070 Ti - 108,5 tok/s Generation, 1.619 tok/s Prefill, TTFT 37.385 ms (3 Laufe)NVIDIA GeForce RTX 50...NVIDIA RTX PRO 6000 Blackwell Workstation Edition - 725,6 tok/s Generation, 8.070 tok/s Prefill, TTFT 3.939 ms (6 Laufe) | DIESER LAUF★ NVIDIA RTX PRO 6000 B...
★ NVIDIA RTX PRO 6000 Blackwell Workstation Edition 725,6 tok/s this runNVIDIA GeForce RTX 3090 Ti 331,6 tok/sNVIDIA GeForce RTX 5070 Ti 108,5 tok/s

CPUby processor

9226914612300,003.2386.4779.715Prefill (tok/s)Generation (tok/s)AMD Ryzen 9 8945HX with Radeon Graphics - 331,6 tok/s Generation, 3.668 tok/s Prefill, TTFT 14.010 ms (3 Laufe)AMD Ryzen 9 8945HX wi...AMD Ryzen Threadripper PRO 5975WX 32-Cores - 108,5 tok/s Generation, 1.619 tok/s Prefill, TTFT 37.385 ms (3 Laufe)AMD Ryzen Threadrippe...AMD Ryzen 9 9950X 16-Core Processor - 725,6 tok/s Generation, 8.070 tok/s Prefill, TTFT 3.939 ms (6 Laufe) | DIESER LAUF★ AMD Ryzen 9 9950X 16-...
★ AMD Ryzen 9 9950X 16-Core Processor 725,6 tok/s this runAMD Ryzen 9 8945HX with Radeon Graphics 331,6 tok/sAMD Ryzen Threadripper PRO 5975WX 32-Cores 108,5 tok/s

MBby mainboard

9226914612300,003.2386.4779.715Prefill (tok/s)Generation (tok/s)Meigao Innovation Technology (Shen Zhen) Co., Ltd DRFXL (MotherBoard Series) - 331,6 tok/s Generation, 3.668 tok/s Prefill, TTFT 14.010 ms (3 Laufe)Meigao Innovation Tec...ASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI - 108,5 tok/s Generation, 1.619 tok/s Prefill, TTFT 37.385 ms (3 Laufe)ASUSTeK COMPUTER INC....ASUSTeK COMPUTER INC. ProArt X870E-CREATOR WIFI - 725,6 tok/s Generation, 8.070 tok/s Prefill, TTFT 3.939 ms (6 Laufe) | DIESER LAUF★ ASUSTeK COMPUTER INC....
★ ASUSTeK COMPUTER INC. ProArt X870E-CREATOR WIFI 725,6 tok/s this runMeigao Innovation Technology (Shen Zhen) Co., Ltd DRFXL (MotherBoard Series) 331,6 tok/sASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI 108,5 tok/s

ENGby engine

8147506856215563.4475.2777.1088.938Prefill (tok/s)Generation (tok/s)vLLM - 725,6 tok/s Generation, 7.865 tok/s Prefill, TTFT 1.602 ms (3 Laufe)vLLMllama.cpp - 645,1 tok/s Generation, 4.521 tok/s Prefill, TTFT 19.223 ms (9 Laufe) | DIESER LAUF★ llama.cpp
vLLM 725,6 tok/s★ llama.cpp 645,1 tok/s this run

DRVby driver

7987627266896535.0355.2505.4645.678Prefill (tok/s)Generation (tok/s)unbekannt - 725,6 tok/s Generation, 5.357 tok/s Prefill, TTFT 14.818 ms (12 Laufe)unbekannt
unbekannt 725,6 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 0.15
Token / kWh2.00M
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)22.60B
☁️ 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

Devstral-Small-2-24B-Instruct-2512NVIDIA RTX PRO 6000 Blackwell Workstation EditionDevstral-Small-2-24B-Instruct-2512NVIDIA RTX PRO 6000 Blackwell Workstation EditionDevstral-Small-2-24B-Instruct-2512NVIDIA GeForce RTX 3090 TiDevstral-Small-2-24B-Instruct-2512NVIDIA 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.