Home / Models / Phi-3 Medium 128K Instruct / NVIDIA RTX 5080 16 GB
Can Phi-3 Medium 128K Instruct run on NVIDIA RTX 5080 16 GB?
Yes. Calculated memory fit: Phi-3 Medium 128K Instruct can run entirely in this GPU’s usable VRAM at one or more supported quantizations.
Can it run?
- Can it run?
- Yes, full GPU fit (calculated estimate).
- Full GPU fit?
- Yes
- Model
- Phi-3 Medium 128K Instruct · 14B
- GPU
- NVIDIA RTX 5080 16 GB · 16 GB advertised
What else can NVIDIA RTX 5080 16 GB run? · Find other GPUs for Phi-3 Medium 128K Instruct
Memory calculation
Breakdown for Q4 at 8K context, batch size 1. Calculator v1.0.0.
| Model weights | 8.2 GB |
|---|---|
| KV cache | 1.6 GB |
| Runtime reserve | 0.9 GB |
| Safety margin | 0.5 GB |
| Required (estimated) | 11.2 GB |
| GPU usable VRAM | 14.4 GB |
Result: FITS IN VRAM
Quantization table
| Quantization | 2K | 4K | 8K | 16K | 32K | 64K | 128K |
|---|---|---|---|---|---|---|---|
| BF16 | Offload | Offload | Offload | Offload | Offload | Offload | Offload |
| FP16 | Offload | Offload | Offload | Offload | Offload | Offload | Offload |
| Q3 | Fits | Fits | Fits | Fits | Fits | Limited context | Limited context |
| Q4 | Fits | Fits | Fits | Fits | Limited context | Limited context | Limited context |
| Q5 | Fits | Fits | Fits | Limited context | Limited context | Limited context | Limited context |
| Q6 | Fits | Fits | Limited context | Limited context | Limited context | Limited context | Limited context |
| Q8 | Offload | Offload | Offload | Offload | Offload | Offload | Offload |
What fits entirely in VRAM?
- Q3 at 16K context
- Q3 at 2K context
- Q3 at 32K context
- Q3 at 4K context
- Q3 at 8K context
- Q4 at 16K context
- Q4 at 2K context
- Q4 at 4K context
- Q4 at 8K context
- Q5 at 2K context
- Q5 at 4K context
- Q5 at 8K context
What requires RAM offload?
Offload is shown only when the model does not fully fit in usable VRAM but stays within the modeled offload allowance.
Performance
Separate from memory fit. Runtime is llama.cpp · quantization bucket Q4 (GGUF Q4_0 maps to Q4). Workload is llama-bench pp512 / tg128 (short context, batch 1, full GPU offload). This is not a ranking and not a 8K–128K speed claim.
- Model
- Phi-3 Medium 128K Instruct
- GPU
- NVIDIA RTX 5080 16 GB
- Runtime
- llama.cpp · CUDA · model 1.0.0
- Decode
- ~80 tok/s · Estimated · MEDIUM confidence (spread ~60 tok/s–~100 tok/s)
- Prefill
- unavailable for this workload
- Time to first token
- unavailable
decode_tps = GPU_efficiency × memory_bandwidth / model_weight_bytes; efficiency from same-gpu Llama 2 7B Q4_0 efficiency; param_ratio=2.08
Available graphics cards
These SKUs use this GPU chip. Amazon CTAs appear only for EXACT/HIGH matches. Absence of a price does not change the compatibility result above.
GIGABYTE GeForce RTX 5080 GAMING OC 16G Graphics Card
16 GB · Amazon EXACT
Check price on Amazon
MSI GAMING GEFORCE RTX 5080 16G TRIO OC graphics card NVIDIA 16 GB GDDR7
16 GB · Amazon EXACT
Check price on Amazon
ASUS Prime -RTX5080-O16G NVIDIA GeForce RTX 5080 16 GB GDDR7
16 GB · Amazon EXACT
Check price on Amazon
ASUS ROG Astral - -RTX5080-O16G-GAMING NVIDIA GeForce RTX 5080 16 GB GDDR7
16 GB · Amazon EXACT
Check price on Amazon
GIGABYTE GeForce RTX 5080 WINDFORCE OC SFF 16G Graphics Card
16 GB · Amazon EXACT
Check price on Amazon
ASUS TUF Gaming TUF-RTX5080-O16G-GAMING NVIDIA GeForce RTX 5080 16 GB GDDR7
16 GB · Amazon EXACT
Check price on Amazon
Sources
Model: official config microsoft/Phi-3-medium-128k-instruct (main). GPU/product specs: Icecat. Compatibility: RigForAI calculator v1.0.0 estimated VRAM · Q8 at 8K: CPU_RAM_OFFLOAD_REQUIRED. Amazon: affiliate commerce match, not a spec source.