Contributed byMario AlkaMistral AI Devstral-Small-2-24B-Instruct-2512
Performance benchmark · measured on 20.07.2026 11:28
Benchmark-ID
run-20260722-165903-d5ed62Dense24BRuntime: godclawQuantisierung: Q8_0
Generation8,88tok/s
Prefill22,07tok/s
Time to First Token906,00ms
Total duration14,99s
Concurrency1parallel
Ranking in the field
17of 19 systems
Performance benchmark · Primary metric: Generation-Speed (tok/s) · 1× concurrent
This run is better than 6 % of all comparable systems.
Generation
8,9 tok/s
-68 % vs Ø 27,5
Prefill
22,1 tok/s
-95 % vs Ø 433,4
Time to First Token
906 ms
-56 % vs Ø 2.049
Distribution in the field7 – 54 tok/s
Wie schlägt sich dieser Benchmark mit anderen Modellen?
Qwen-AgentWorld-35B-A3BAMD Radeon 8060S Graphics · run-20260723-142108-f59a48
53,7 tok/s
Gemma-4-26B-A4B-itAMD Radeon 8060S Graphics · run-20260722-165910-670419
42,2 tok/s
Gemma-4-26B-A4B-itAMD Radeon 8060S Graphics · run-20260722-165909-f81e9d
42,2 tok/s
Gemma-4-26B-A4B-itAMD Radeon 8060S Graphics · run-20260723-023424-7d69dd
39,5 tok/s
gemma-4-26B-A4B-it-assistantAMD Radeon 8060S Graphics · run-20260725-031727-ec1196
33,0 tok/s
Gemma-4-26B-A4B-itAMD Radeon 8060S Graphics · run-20260725-031733-4dcb5c
32,8 tok/s
Gemma-4-26B-A4B-itAMD Radeon 8060S Graphics · run-20260725-031733-216603
32,4 tok/s
Qwen3.6-35B-A3BAMD Radeon 8060S Graphics · run-20260818-172656-f86118
27,7 tok/s
Qwen3.6-35B-A3BAMD Radeon 8060S Graphics · run-20260820-184209-9c541a
27,6 tok/s
Qwen3.6-35B-A3BAMD Radeon 8060S Graphics · run-20260818-172656-c87b94
27,6 tok/s
Nemotron-Cascade-2-30B-A3BAMD Radeon 8060S Graphics · run-20260820-184209-d64192
27,5 tok/s
Nemotron-Cascade-2-30B-A3BAMD Radeon 8060S Graphics · run-20260820-184209-dca769
27,5 tok/s
Qwen3.6-35B-A3BAMD Radeon 8060S Graphics · run-20260821-053525-989728
26,9 tok/s
Qwen3.6-35B-A3BAMD Radeon 8060S Graphics · run-20260818-174731-31bdeb
24,5 tok/s
Devstral-Small-2-24B-Instruct-2512 this runAMD Radeon 8060S Graphics · run-20260722-165903-d5ed62
8,9 tok/s
How does this benchmark compare on other GPUs?
Same model on different hardware · 1× concurrent · Generation (tok/s)
NVIDIA RTX PRO 6000 Blackwell Workstation Editionrun-20260724-004603-630e7a
96,9 tok/s
NVIDIA GeForce RTX 5090run-20260729-105333-aaec90
94,6 tok/s
NVIDIA RTX PRO 6000 Blackwell Workstation Editionrun-20260729-032117-6b0bb2
94,5 tok/s
3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Editionrun-20260729-105337-2f5f33
88,7 tok/s
3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Editionrun-20260729-105337-72ec92
88,6 tok/s
3× NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Editionrun-20260729-105337-9999aa
80,7 tok/s
NVIDIA GeForce RTX 3090 Tirun-20260729-032129-c03400
56,2 tok/s
AMD Radeon PRO W7900 Dual Slotrun-20260817-212318-ab5bc0
47,8 tok/s
2× AMD Radeon PRO W7900 Dual Slotrun-20260818-060834-ed6b7f
47,3 tok/s
2× AMD Radeon PRO W7900 Dual Slotrun-20260818-064050-19d39a
38,3 tok/s
3× AMD Radeon AI PRO R9700run-20260805-053401-cf14d0
32,6 tok/s
3× AMD Radeon AI PRO R9700run-20260805-053401-efbc33
32,3 tok/s
3× AMD Radeon AI PRO R9700run-20260805-053402-6aced6
31,9 tok/s
3× AMD Radeon AI PRO R9700run-20260722-165907-d98515
29,4 tok/s
Hardware
GPU: AMD Radeon 8060S Graphics
CPU: 32x AMD RYZEN AI MAX+ 395 w/ Radeon 8060S
RAM: 31 GB
Mainboard: Meigao Innovation Technology (Shen Zhen) Co., Ltd SHWSA (MS-S1 MAX)
Setup
Runtime: godclaw
Quantization: Q8_0
Operating system: Ubuntu 24.04.2 LTS (Kernel 6.17.0-40-generic)
Driver: ROCm 7.2.0
Model: Devstral-Small-2-24B-Instruct-2512
Configuration
# LLM-Benchmark Konfiguration
# Modell : Devstral-Small-2-24B-Instruct-2512
# Run-ID : run-20260722-165903-d5ed62
# GPU : AMD Radeon 8060S Graphics
# CPU : 32x AMD RYZEN AI MAX+ 395 w/ Radeon 8060S
# RAM : 31 GB
bench@llm-benchmark:~$ cat benchmark.conf
Konfigurationspfad /home/godcore/models/devstral-small-2-24b-q8_0-gguf/Devstral-Small-2-24B-Instruct-2512-Q8_0.gguf
Engine llamacpp
Modellalias mistralai/Devstral-Small-2-24B-Instruct-2512
Kontextlaenge 131072
Host 0.0.0.0
Port 8000
Parallel 1
Temperatur 0.2
Repeat-Penalty 1.05
Repeat-Last-N 256
GPU-Layer 99
Flash Attention on
Jinja aktiv
Chat-Template /home/godcore/models/devstral-small-2-24b-q8_0-gguf/chat_template.jinja
Threads 8
Batch-Threads 16
| 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. | mistralai/Devstral-Small-2-24B-Instruct-2512 |
| Kontextlaenge?Maximale Anzahl Tokens (Eingabe + erzeugte Ausgabe zusammen), die das Modell pro Anfrage verarbeiten kann. | 131072 |
| Modellpfad?Pfad zur GGUF-Modelldatei, die geladen und ausgeliefert wird. | /home/godcore/models/devstral-small-2-24b-q8_0-gguf/Devstral-Small-2-24B-Instruct-2512-Q8_0.gguf |
| Alias?Anzeigename des Modells nach aussen (served model name), unabhaengig vom Dateinamen. | mistralai/Devstral-Small-2-24B-Instruct-2512 |
| Host?Netzwerk-Interface, an das der HTTP-Server bindet, z.B. 0.0.0.0 fuer alle Interfaces. | 0.0.0.0 |
| Port?TCP-Port des HTTP-Servers. | 8000 |
| Kontext?Groesse des Kontextfensters in Token. 0 = der beim Training verwendete Kontext des Modells. | 131072 |
| Parallel?Anzahl paralleler Slots/Sequenzen, die der Server gleichzeitig bedient. Der Kontext wird auf die Slots aufgeteilt. | 1 |
| Temperatur?Sampling-Temperatur. Hoeher = kreativer/zufaelliger, niedriger = deterministischer. 0 = greedy. | 0.2 |
| Repeat-Penalty?Straffaktor gegen Token-Wiederholungen. 1.0 = keine Strafe, hoeher = staerker. | 1.05 |
| Repeat-Last-N?Anzahl der letzten Token, die fuer die Wiederholungsstrafe betrachtet werden. 0 = aus, -1 = gesamter Kontext. | 256 |
| 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. | 99 |
| Flash Attention?FlashAttention fuer schnellere und speichersparende Attention. Wert on/off/auto je nach Build. | on |
| Jinja?Nutzt die im Modell eingebettete Jinja-Chat-Vorlage fuer korrekte Rollen-, Tool- und Reasoning-Formatierung. | aktiv |
| Chat-Template?Pfad zu einer externen Chat-Template-Datei. Ueberschreibt die im Modell eingebettete Vorlage. | /home/godcore/models/devstral-small-2-24b-q8_0-gguf/chat_template.jinja |
| Threads?Anzahl CPU-Threads fuer die Token-Generierung (Decode). | 8 |
| Batch-Threads?Anzahl CPU-Threads fuer Prompt-Verarbeitung und Batch (Prefill). | 16 |
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
NVIDIA RTX PRO 6000 Blackwell Workstation Edition 725,6 tok/sNVIDIA GeForce RTX 5090 620,3 tok/sNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition 569,7 tok/sNVIDIA GeForce RTX 3090 Ti 331,6 tok/sAMD Radeon AI PRO R9700 206,3 tok/sAMD Radeon PRO W7900 Dual Slot 152,1 tok/sNVIDIA GeForce RTX 5070 Ti 108,5 tok/sNVIDIA Tesla P100 PCIe 16GB 23,3 tok/s★ AMD Radeon 8060S Graphics 8,9 tok/s this run
CPUby processor
AMD Ryzen 9 9950X 16-Core Processor 725,6 tok/sAMD Ryzen 7 5800X3D 8-Core Processor 620,3 tok/sAMD Ryzen Threadripper PRO 9965WX 24-Cores 569,7 tok/sAMD Ryzen 9 8945HX with Radeon Graphics 331,6 tok/sAMD Ryzen 5 5600X 6-Core Processor 247,7 tok/sAMD Ryzen Threadripper PRO 7955WX 16-Cores 206,3 tok/sAMD Ryzen Threadripper PRO 5975WX 32-Cores 152,1 tok/sAMD Ryzen 9 7945HX with Radeon Graphics 23,3 tok/s★ AMD RYZEN AI MAX+ 395 w/ Radeon 8060S 8,9 tok/s this run
MBby mainboard
ASUSTeK COMPUTER INC. ProArt X870E-CREATOR WIFI 725,6 tok/sASUSTeK COMPUTER INC. ROG STRIX B550-A GAMING 620,3 tok/sASUSTeK COMPUTER INC. Pro WS WRX90E-SAGE SE 569,7 tok/sMeigao Innovation Technology (Shen Zhen) Co., Ltd DRFXL (MotherBoard Series) 331,6 tok/sASUSTeK COMPUTER INC. PRIME A520M-K 247,7 tok/sASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI 152,1 tok/sShenzhen Meigao Electronic Equipment Co.,Ltd F1FXM (DeskMini Series) 23,3 tok/s★ Meigao Innovation Technology (Shen Zhen) Co., Ltd SHWSA (MS-S1 MAX) 8,9 tok/s this runBosgame AXB35-02 (BeyondMax Series) 8,9 tok/s
DRVby driver
unbekannt 725,6 tok/sAMD 7.0.0-27-generic 29,4 tok/s★ ROCm 7.2.0 8,9 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 (1× 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
18 W
⚡ TDP
120 W
☁️ External LLM (API)
Electricity0.30 EUR/kWh
Avg power (incl. idle)120 W estimated (TDP)GPU 120 W full load
Avg cost / hourEUR 0.036
Electricity / 1M tokensEUR 1.13
Token / kWh266.40K
Acquisition (system)EUR 4,454 missingBoard EUR 3,900 · RAM EUR 434 · PSU EUR 120
Electricity (2 years)–
TCO (2 years)EUR 5,085
Output tokens (2 years)560.08M
☁️ 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 (18 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-2512AMD Radeon 8060S Graphics | Devstral-Small-2-24B-Instruct-2512NVIDIA RTX PRO 6000 Blackwell Workstation Edition | Devstral-Small-2-24B-Instruct-2512NVIDIA GeForce RTX 5090 | Devstral-Small-2-24B-Instruct-2512NVIDIA RTX PRO 6000 Blackwell Workstation Edition | |
|---|---|---|---|---|
| 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) | – | – | – | – |
AMD Radeon 8060S Graphics · this run 9 tok/sNVIDIA RTX PRO 6000 Blackwell Workstation Edition 97 tok/sNVIDIA GeForce RTX 5090 95 tok/sNVIDIA RTX PRO 6000 Blackwell Workstation Edition 94 tok/s
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.