Mixtral-8x22B-Instruct (MoE) VRAM Requirements & GPU Benchmarks
Mixtral-8x22B-Instruct (MoE) has 141B total parameters (39B active per token), 56 layers, and 8 KV heads (MoE-GQA). It requires 90.46 GB VRAM in 4-bit (Q4_K_M), 156.72 GB in 8-bit, and 292.87 GB in FP16 at an 8K context window.
CLOUD GPU ALTERNATIVE
Need 90.46 GB+ VRAM to run Mixtral-8x22B-Instruct (MoE)?
Recommended cloud setup for 4-bit: Multi-GPU H100 / H200 Node (or 292.87 GB for full FP16). Spin up a pre-configured vLLM / Ollama instance in 60 seconds:
| Provider | Best For | RTX 4090 (24GB) | A100 (80GB) | H100 (80GB) | Bonus / Trial | Action |
|---|---|---|---|---|---|---|
| RunPod | 1-Click vLLM / Ollama Pods | $0.44 / hr | $1.39 / hr | $2.39 / hr | Instant PyTorch / vLLM Pods + Serverless GPU | Deploy on RunPod → |
| Vast.ai | Cheapest Spot GPU Market | $0.29 / hr | $0.95 / hr | $1.79 / hr | Save 50–70% on RTX 4090 / 5090 / A100 rentals | Check Spot GPUs → |
| Vultr Cloud GPU | Enterprise SLA & Bare Metal | — | $1.79 / hr | $2.69 / hr | $300 Free Credit for New Users | Claim $300 Credit → |
| DigitalOcean GPU Droplets | Managed H100 Droplets | — | — | $2.99 / hr | $200 Free Trial (60 Days) | Get $200 Free → |
GPU Compatibility Matrix for Mixtral-8x22B-Instruct (MoE)
| GPU / Hardware | VRAM | Bandwidth | Fits FP16 (292.87GB)? | Fits Q4_K_M (90.46GB)? | Est. Speed (Q4_K_M) | Detailed Report |
|---|---|---|---|---|---|---|
| NVIDIA RTX 5090 | 32 GB | 1792 GB/s | No | No | ~6.1 tok/s | View NVIDIA RTX 5090 Setup → |
| NVIDIA RTX 4090 | 24 GB | 1008 GB/s | No | No | ~3.4 tok/s | View NVIDIA RTX 4090 Setup → |
| NVIDIA RTX 5080 | 16 GB | 960 GB/s | No | No | ~3.2 tok/s | View NVIDIA RTX 5080 Setup → |
| NVIDIA RTX 4080 Super | 16 GB | 736 GB/s | No | No | ~2.5 tok/s | View NVIDIA RTX 4080 Super Setup → |
| NVIDIA RTX 4070 Ti Super | 16 GB | 672 GB/s | No | No | ~2.3 tok/s | View NVIDIA RTX 4070 Ti Super Setup → |
| NVIDIA RTX 3090 | 24 GB | 936 GB/s | No | No | ~3.2 tok/s | View NVIDIA RTX 3090 Setup → |
| NVIDIA RTX 4060 Ti 16GB | 16 GB | 288 GB/s | No | No | ~1.2 tok/s | View NVIDIA RTX 4060 Ti 16GB Setup → |
| NVIDIA RTX 3060 12GB | 12 GB | 360 GB/s | No | No | ~1.2 tok/s | View NVIDIA RTX 3060 12GB Setup → |
| NVIDIA RTX 6000 Ada | 48 GB | 960 GB/s | No | No | ~3.2 tok/s | View NVIDIA RTX 6000 Ada Setup → |
| NVIDIA RTX A6000 | 48 GB | 768 GB/s | No | No | ~2.6 tok/s | View NVIDIA RTX A6000 Setup → |
| NVIDIA L40S 48GB | 48 GB | 864 GB/s | No | No | ~2.9 tok/s | View NVIDIA L40S 48GB Setup → |
| NVIDIA A100 80GB SXM | 80 GB | 2039 GB/s | No | No | ~6.9 tok/s | View NVIDIA A100 80GB SXM Setup → |
| NVIDIA H100 80GB SXM | 80 GB | 3350 GB/s | No | No | ~11.3 tok/s | View NVIDIA H100 80GB SXM Setup → |
| NVIDIA H200 141GB SXM | 141 GB | 4800 GB/s | No | Yes (64%) | ~125.9 tok/s | View NVIDIA H200 141GB SXM Setup → |
| NVIDIA Blackwell B200 192GB | 192 GB | 8000 GB/s | No | Yes (47%) | ~209.8 tok/s | View NVIDIA Blackwell B200 192GB Setup → |
| AMD Instinct MI300X 192GB | 192 GB | 5300 GB/s | No | Yes (47%) | ~139 tok/s | View AMD Instinct MI300X 192GB Setup → |
| 2x NVIDIA RTX 3090 (48GB) | 48 GB | 936 GB/s | No | No | ~3.2 tok/s | View 2x NVIDIA RTX 3090 (48GB) Setup → |
| 2x NVIDIA RTX 4090 (48GB) | 48 GB | 1008 GB/s | No | No | ~3.4 tok/s | View 2x NVIDIA RTX 4090 (48GB) Setup → |
| 4x NVIDIA A100 80GB (320GB) | 320 GB | 2039 GB/s | Yes | Yes (28%) | ~53.5 tok/s | View 4x NVIDIA A100 80GB (320GB) Setup → |
| 8x NVIDIA H100 80GB (640GB) | 640 GB | 3350 GB/s | Yes | Yes (14%) | ~87.8 tok/s | View 8x NVIDIA H100 80GB (640GB) Setup → |
| Apple M4 Max (128GB Unified) | 128 GB | 546 GB/s | No | Yes (71%) | ~15.7 tok/s | View Apple M4 Max (128GB Unified) Setup → |
| Apple M3 Ultra Mac Studio (512GB) | 512 GB | 819 GB/s | Yes | Yes (18%) | ~23.6 tok/s | View Apple M3 Ultra Mac Studio (512GB) Setup → |
| Apple M4 Pro Mac Mini (64GB) | 64 GB | 273 GB/s | No | No | ~1.2 tok/s | View Apple M4 Pro Mac Mini (64GB) Setup → |