✅ RUNS NATIVE FP16 & ALL QUANTIZATIONS
Microsoft Phi-4 14B on 2x NVIDIA RTX 3090 (48GB) (48GB VRAM)
In 4-bit Q4_K_M, Microsoft Phi-4 14B uses only 11.4 GB of your 48 GB VRAM (24% utilization) at ~65.1 tokens/sec. In full 16-bit precision, it requires 32.5 GB.
🎛️ Interactive VRAM & KV-Cache Tuner
📊 Real-Time Memory Footprint
11.4 GB / 48 GB
🟦 Weights (9.18 GB)
🟪 KV Cache (1.56 GB)
🟨 Activation Buffer (0.66 GB)
Total Required VRAM11.4 GB
Available Hardware VRAM48 GB (2x NVIDIA RTX 3090 (48GB))
VRAM Headroom / Overflow36.60 GB
Estimated Generation Speed~65.1 tokens/sec
Architecture & AttentionDense-GQA (40L, 10 KV heads)
🚀 Ready-to-Run Launch Command
ollama run phi4:14b # Or with llama.cpp server (8K context, Flash Attention): llama-server -hf microsoft/phi-4-GGUF -c 8192 -ngl 99 -fa
CLOUD GPU ALTERNATIVE
Need 11.4 GB+ VRAM to run Microsoft Phi-4 14B?
Recommended cloud setup for 4-bit: RTX 4090 (24GB) (or 32.5 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 → |
All Quantization Formats: Microsoft Phi-4 14B on 2x NVIDIA RTX 3090 (48GB)
| Quantization Format | Bits/Weight | Weights VRAM | KV Cache (8K) | Total VRAM | Fits on 1x 2x NVIDIA RTX 3090 (48GB) (48GB)? | Est. Speed | GPUs Needed |
|---|---|---|---|---|---|---|---|
| FP16 / BF16 (Unquantized) | 16 bpw | 30.28 GB | 1.56 GB | 32.5 GB | Yes (68%) | ~19.7 tok/s | 1x 2x NVIDIA RTX 3090 (48GB) |
| FP8 (W8A8 / E4M3) | 8.5 bpw | 16.09 GB | 1.56 GB | 18.31 GB | Yes (38%) | ~37.2 tok/s | 1x 2x NVIDIA RTX 3090 (48GB) |
| GGUF Q8_0 (8-bit) | 8.5 bpw | 16.09 GB | 1.56 GB | 18.31 GB | Yes (38%) | ~37.2 tok/s | 1x 2x NVIDIA RTX 3090 (48GB) |
| GGUF Q6_K (6-bit) | 6.56 bpw | 12.42 GB | 1.56 GB | 14.64 GB | Yes (31%) | ~48.1 tok/s | 1x 2x NVIDIA RTX 3090 (48GB) |
| GGUF Q5_K_M (5-bit) | 5.69 bpw | 10.77 GB | 1.56 GB | 12.99 GB | Yes (27%) | ~55.5 tok/s | 1x 2x NVIDIA RTX 3090 (48GB) |
| GGUF Q4_K_M / AWQ (4-bit) | 4.85 bpw | 9.18 GB | 1.56 GB | 11.4 GB | Yes (24%) | ~65.1 tok/s | 1x 2x NVIDIA RTX 3090 (48GB) |
| GGUF IQ4_XS (4.25 bpw) | 4.25 bpw | 8.04 GB | 1.56 GB | 10.27 GB | Yes (21%) | ~74.3 tok/s | 1x 2x NVIDIA RTX 3090 (48GB) |
| GGUF Q3_K_M (3-bit) | 3.91 bpw | 7.4 GB | 1.56 GB | 9.62 GB | Yes (20%) | ~80.8 tok/s | 1x 2x NVIDIA RTX 3090 (48GB) |
Frequently Asked Questions (Microsoft Phi-4 14B + 2x NVIDIA RTX 3090 (48GB))
Can 2x NVIDIA RTX 3090 (48GB) (48GB) run Microsoft Phi-4 14B locally?
Yes — 2x NVIDIA RTX 3090 (48GB) (48GB) easily runs Microsoft Phi-4 14B even in full FP16. In 4-bit Q4_K_M, Microsoft Phi-4 14B uses only 11.4 GB of your 48 GB VRAM (24% utilization) at ~65.1 tokens/sec. In full 16-bit precision, it requires 32.5 GB.
How is KV Cache calculated for Microsoft Phi-4 14B?
Microsoft Phi-4 14B uses Dense-GQA with 40 transformer layers, 40 attention heads, and 10 KV heads (head dimension 128). At 8,192 tokens in FP16, KV cache consumes 1.56 GB. Enabling 8-bit KV cache (-ctk q8_0 -ctv q8_0 in llama.cpp or --kv-cache-dtype fp8 in vLLM) cuts this in half to 0.78 GB.
Compare Microsoft Phi-4 14B on Other GPUs
Microsoft Phi-4 14B on NVIDIA RTX 5090 (32GB)Microsoft Phi-4 14B on NVIDIA RTX 4090 (24GB)Microsoft Phi-4 14B on NVIDIA RTX 5080 (16GB)Microsoft Phi-4 14B on NVIDIA RTX 4080 Super (16GB)Microsoft Phi-4 14B on NVIDIA RTX 4070 Ti Super (16GB)Microsoft Phi-4 14B on NVIDIA RTX 3090 (24GB)Microsoft Phi-4 14B on NVIDIA RTX 4060 Ti 16GB (16GB)Microsoft Phi-4 14B on NVIDIA RTX 3060 12GB (12GB)Microsoft Phi-4 14B on NVIDIA RTX 6000 Ada (48GB)Microsoft Phi-4 14B on NVIDIA RTX A6000 (48GB)
Other Models on 2x NVIDIA RTX 3090 (48GB) (48GB)
DeepSeek-R1 671B (MoE) on 2x NVIDIA RTX 3090 (48GB)DeepSeek-V3 671B (MoE) on 2x NVIDIA RTX 3090 (48GB)DeepSeek-R1-Distill-Llama-70B on 2x NVIDIA RTX 3090 (48GB)DeepSeek-R1-Distill-Qwen-32B on 2x NVIDIA RTX 3090 (48GB)DeepSeek-R1-Distill-Qwen-14B on 2x NVIDIA RTX 3090 (48GB)DeepSeek-R1-Distill-Llama-8B on 2x NVIDIA RTX 3090 (48GB)DeepSeek-R1-Distill-Qwen-7B on 2x NVIDIA RTX 3090 (48GB)Qwen3-235B-A22B (MoE) on 2x NVIDIA RTX 3090 (48GB)Qwen3-30B-A3B (MoE) on 2x NVIDIA RTX 3090 (48GB)Qwen3-32B on 2x NVIDIA RTX 3090 (48GB)