✅ LÄUFT NATIV IN FP16 & ALLEN QUANTISIERUNGEN
Mistral-Large-2-123B on 8x NVIDIA H100 80GB (640GB) (640GB VRAM)
In 4-bit GGUF Q4_K_M, Mistral-Large-2-123B consumes 72.84 GB (69.59 GB weights + 2.75 GB KV cache at 8K context), achieving ~31.1 tokens/sec decode and ~16779 tokens/sec prefill.
🎛️ Interaktiver VRAM-, Kontext- & Offload-Konfigurator
📊 Echtzeit-Speicher- & Performance-Telemetrie
72.84 GB / 640 GB
🟦 Modellgewichte: 69.59 GB
🟪 KV-Cache: 2.75 GB
🟨 Engine-Puffer: 0.51 GB
Benötigter GPU-VRAM72.84 GB
Benötigter System-RAM (CPU)0.00 GB
Verfügbarer Hardware-VRAM640 GB (1x 8x NVIDIA H100 80GB (640GB))
VRAM-Reserve / Defizit567.16 GB
Geschätzte Decode-Geschwindigkeit~31.1 tokens/sec
Geschätzte Prefill-Geschwindigkeit~16779 tokens/sec
Time-to-First-Token (4K Prompt)~244 ms
Transformer-TopologieDense-GQA (88L, 8/96 KV/Q heads)
🚀 Startbefehl (vLLM / llama.cpp / Ollama)
ollama run mistral-large:123b # Or llama.cpp server (8K context, Flash Attention): llama-server -hf mistralai/Mistral-Large-Instruct-2411-GGUF -c 8192 -ngl 99 -fa
CLOUD-GPU PREISVERGLEICH & MIETEN-VS-KAUFEN RECHNER
Nicht genug lokaler VRAM? Cloud-GPU in 60 Sekunden starten (Mistral-Large-2-123B: 72.84 GB in 4-bit)
Recommended Cloud Tier: NVIDIA A100 / H100 (80GB) · Full FP16 Needs: 232.07 GB
Amortisationszeit (Kauf vs. Miete): ~32 months (vs $230,000 hardware)
| Cloud Provider | RTX 4090 (24GB) | A100 (80GB) | H100 (80GB) | Est. Monthly Cost (4h/day) | Free Credit / Trial | Deploy |
|---|---|---|---|---|---|---|
| RunPod | $0.44 / hr | $1.39 / hr | $2.39 / hr | $52.80 / mo | Instant PyTorch / vLLM Templates + Serverless GPU | Auf RunPod starten → |
| Vast.ai | $0.29 / hr | $0.95 / hr | $1.79 / hr | $34.80 / mo | Save 50–70% on RTX 4090 / 5090 / A100 rentals | Spot-GPU mieten → |
| Vultr Cloud GPU | — | $1.79 / hr | $2.69 / hr | $214.80 / mo (A100) | $300 Free Credit | $300 Guthaben sichern → |
| DigitalOcean GPU | — | — | $2.99 / hr | $358.80 / mo (H100) | $200 Free Credit (60d) | $200 Startguthaben → |
Alle Quantisierungsstufen im Detail: Mistral-Large-2-123B × 8x NVIDIA H100 80GB (640GB)
| Quantization Format | Core / Emb BPW | Weights (GiB) | KV Cache (8K) | Total VRAM | 1x 8x NVIDIA H100 80GB (640GB) (640GB) | Decode / Prefill | Min GPUs |
|---|---|---|---|---|---|---|---|
| FP16 / BF16 (Unquantized) | 16 / 16 | 228.36 GB | 2.75 GB | 231.62 GB | Ja (36%) | ~9.4 / 16779 tok/s | 1x |
| FP8 (W8A8 / E4M3) | 8.25 / 8.5 | 117.84 GB | 2.75 GB | 121.1 GB | Ja (19%) | ~18.2 / 16779 tok/s | 1x |
| GGUF Q8_0 (8.50 bpw) | 8.5 / 8.5 | 121.32 GB | 2.75 GB | 124.57 GB | Ja (19%) | ~17.7 / 16779 tok/s | 1x |
| GGUF Q6_K (6.56 bpw) | 6.56 / 6.56 | 93.63 GB | 2.75 GB | 96.88 GB | Ja (15%) | ~22.9 / 16779 tok/s | 1x |
| GGUF Q5_K_M (5.69 bpw) | 5.69 / 6.56 | 81.54 GB | 2.75 GB | 84.79 GB | Ja (13%) | ~26.4 / 16779 tok/s | 1x |
| GGUF Q4_K_M / AWQ (4.83 bpw) | 4.83 / 6.56 | 69.59 GB | 2.75 GB | 72.84 GB | Ja (11%) | ~31.1 / 16779 tok/s | 1x |
| GGUF IQ4_XS (4.25 bpw) | 4.25 / 5.5 | 61.13 GB | 2.75 GB | 64.38 GB | Ja (10%) | ~35.3 / 16779 tok/s | 1x |
| GGUF Q3_K_M (3.91 bpw) | 3.91 / 5 | 56.21 GB | 2.75 GB | 59.47 GB | Ja (9%) | ~38.4 / 16779 tok/s | 1x |
Dieses Modell auf anderen GPUs testen
Mistral-Large-2-123B on NVIDIA RTX 5090 (32GB)Mistral-Large-2-123B on NVIDIA RTX 4090 (24GB)Mistral-Large-2-123B on NVIDIA RTX 5080 (16GB)Mistral-Large-2-123B on NVIDIA RTX 4080 Super (16GB)Mistral-Large-2-123B on NVIDIA RTX 4070 Ti Super (16GB)Mistral-Large-2-123B on NVIDIA RTX 3090 (24GB)Mistral-Large-2-123B on NVIDIA RTX 4060 Ti 16GB (16GB)Mistral-Large-2-123B on NVIDIA RTX 3060 12GB (12GB)Mistral-Large-2-123B on NVIDIA RTX 6000 Ada (48GB)Mistral-Large-2-123B on NVIDIA RTX A6000 (48GB)
Weitere Modelle auf dieser GPU
DeepSeek-R1 671B (MoE) on 8x NVIDIA H100 80GB (640GB)DeepSeek-V3-0324 671B (MoE) on 8x NVIDIA H100 80GB (640GB)DeepSeek-R1-Distill-Llama-70B on 8x NVIDIA H100 80GB (640GB)DeepSeek-R1-Distill-Qwen-32B on 8x NVIDIA H100 80GB (640GB)DeepSeek-R1-Distill-Qwen-14B on 8x NVIDIA H100 80GB (640GB)DeepSeek-R1-Distill-Llama-8B on 8x NVIDIA H100 80GB (640GB)DeepSeek-R1-Distill-Qwen-7B on 8x NVIDIA H100 80GB (640GB)Qwen3-235B-A22B (MoE) on 8x NVIDIA H100 80GB (640GB)Qwen3-30B-A3B (MoE) on 8x NVIDIA H100 80GB (640GB)Qwen3-32B on 8x NVIDIA H100 80GB (640GB)