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

Mamba-Codestral-7B-v0.1

Performance benchmark · measured on 28.07.2026 19:27

Benchmark-IDrun-20260728-194136-498f6c
Timebench 3 - Kombi (Prefill + Generation)Dense7BRuntime: llama.cppQuantisierung: Q4_K_M
Generation253,40tok/s
Prefill11.564,80tok/s
Time to First Token234,50ms
Total duration8,55s
Concurrency1parallel
Ranking in the field
8of 38 systems

Performance benchmark · Primary metric: Generation-Speed (tok/s) · 1× concurrent

This run is better than 81 % of all comparable systems.
Generation 253,4 tok/s
+52 % vs Ø 166,2
Prefill 11.564,8 tok/s
+121 % vs Ø 5.229,5
Time to First Token 235 ms
-93 % vs Ø 3.376
Distribution in the field0 – 393 tok/s
Ø 166 Median Dieser Lauf

Wie schlägt sich dieser Benchmark mit anderen Modellen?

NVIDIA-Nemotron-3-Nano-4BNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260727-181607-bffb11
393,5 tok/s
gemma-4-E2B-itNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260727-181606-6ea089
378,6 tok/s
Laguna-XS-2.1NVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260728-140954-cecd39
310,2 tok/s
Laguna-XS-2.1NVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260728-140955-42efc5
309,9 tok/s
Laguna-XS-2.1NVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260728-140955-abb113
309,5 tok/s
North-Mini-Code-1.0NVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260728-164528-7e5f02
274,4 tok/s
gpt-oss-20bNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260723-194447-636902
268,7 tok/s
Mamba-Codestral-7B-v0.1 this runNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260728-194136-498f6c
253,4 tok/s
gemma-4-E4B-itNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260727-181607-d807ba
246,5 tok/s
Qwen3-Coder-NextNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260728-184455-777e64
242,2 tok/s
Meta-Llama-3.1-8B-InstructNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260728-194136-f84bc1
239,3 tok/s
Qwen3-Coder-30B-A3B-InstructNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260724-004605-d79930
237,2 tok/s
Qwen3-30B-A3B-Thinking-2507NVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260724-004605-ffa4c3
233,9 tok/s
Qwen3-Omni-30B-A3B-ThinkingNVIDIA RTX PRO 6000 Blackwell Workstation Edition · run-20260724-004611-89d26d
229,0 tok/s

How does this benchmark compare on other GPUs?

Same model on different hardware · 1× 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: Mamba-Codestral-7B-v0.1

Configuration

benchmark-konfiguration — run-20260728-194136-498f6c
# LLM-Benchmark Konfiguration # Modell : Mamba-Codestral-7B-v0.1 # Engine : llama.cpp # Run-ID : run-20260728-194136-498f6c # 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--Agnuxo--Mamba-Codestral-7B-v0.1-instruct-python_coding_assistant-GGUF_4bit/snapshots/bff85ef20c6077b229e92fcc4eb7cb15e619e9f6/unsloth.Q4_K_M.gguf \ --alias Mamba-Codestral-7B-v0.1 \ --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.Mamba-Codestral-7B-v0.1
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--Agnuxo--Mamba-Codestral-7B-v0.1-instruct-python_coding_assistant-GGUF_4bit/snapshots/bff85ef20c6077b229e92fcc4eb7cb15e619e9f6/unsloth.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

Mamba-Codestral-7B-v0.1 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

2.0331.6871.3419946484.4119.50114.59119.682Prefill (tok/s)Generation (tok/s)NVIDIA GeForce RTX 5070 Ti - 968,5 tok/s Generation, 7.178 tok/s Prefill, TTFT 5.296 ms (3 Laufe)NVIDIA GeForce RTX 50...NVIDIA RTX PRO 6000 Blackwell Workstation Edition - 1.712,6 tok/s Generation, 16.914 tok/s Prefill, TTFT 2.748 ms (3 Laufe) | DIESER LAUF★ NVIDIA RTX PRO 6000 B...
★ NVIDIA RTX PRO 6000 Blackwell Workstation Edition 1.712,6 tok/s this runNVIDIA GeForce RTX 5070 Ti 968,5 tok/s

CPUby processor

2.0331.6871.3419946484.4119.50114.59119.682Prefill (tok/s)Generation (tok/s)AMD Ryzen Threadripper PRO 5975WX 32-Cores - 968,5 tok/s Generation, 7.178 tok/s Prefill, TTFT 5.296 ms (3 Laufe)AMD Ryzen Threadrippe...AMD Ryzen 9 9950X 16-Core Processor - 1.712,6 tok/s Generation, 16.914 tok/s Prefill, TTFT 2.748 ms (3 Laufe) | DIESER LAUF★ AMD Ryzen 9 9950X 16-...
★ AMD Ryzen 9 9950X 16-Core Processor 1.712,6 tok/s this runAMD Ryzen Threadripper PRO 5975WX 32-Cores 968,5 tok/s

MBby mainboard

2.0331.6871.3419946484.4119.50114.59119.682Prefill (tok/s)Generation (tok/s)ASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI - 968,5 tok/s Generation, 7.178 tok/s Prefill, TTFT 5.296 ms (3 Laufe)ASUSTeK COMPUTER INC....ASUSTeK COMPUTER INC. ProArt X870E-CREATOR WIFI - 1.712,6 tok/s Generation, 16.914 tok/s Prefill, TTFT 2.748 ms (3 Laufe) | DIESER LAUF★ ASUSTeK COMPUTER INC....
★ ASUSTeK COMPUTER INC. ProArt X870E-CREATOR WIFI 1.712,6 tok/s this runASUSTeK COMPUTER INC. Pro WS WRX80E-SAGE SE WIFI 968,5 tok/s

ENGby engine

1.8841.7981.7131.6271.54111.32311.80512.28712.769Prefill (tok/s)Generation (tok/s)llama.cpp - 1.712,6 tok/s Generation, 12.046 tok/s Prefill, TTFT 4.022 ms (6 Laufe) | DIESER LAUF★ llama.cpp
★ llama.cpp 1.712,6 tok/s this run

DRVby driver

1.8841.7981.7131.6271.54111.32311.80512.28712.769Prefill (tok/s)Generation (tok/s)unbekannt - 1.712,6 tok/s Generation, 12.046 tok/s Prefill, TTFT 4.022 ms (6 Laufe)unbekannt
unbekannt 1.712,6 tok/s
💰 Economics

Economics of this run

Operating cost, TCO and comparison with the next-best runs of the same model at identical concurrency (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 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 0.21
Token / kWh1.42M
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)15.98B
☁️ 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

Mamba-Codestral-7B-v0.1NVIDIA RTX PRO 6000 Blackwell Workstation EditionMamba-Codestral-7B-v0.1NVIDIA 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.