Can the NVIDIA RTX 4090 run Phi-3.5 Mini 3.8B?

RTX 4090 × Phi-3.5 Mini 3.8B8K context
YesRuns fully on the GPU
Memory at FP1611.5 GB of 24 GB
Weights 7.6 GBKV cache 3.2 GBRuntime 0.6 GB
QuantFP16Full quality
Memory11.5 GBof 24 GB usable
Speed~79tokens per second
  1. FP16
  2. Q8_0
  3. Q6_K
  4. Q5_K_M
  5. Q4_K_M
  6. Q3_K_M
  7. Q2_K

Yes. The NVIDIA RTX 4090 can run Phi-3.5 Mini 3.8B fully in VRAM at FP16 (full quality), using about 11.5 GB of the 24 GB available, at roughly 79 tokens/second.

Run it with Ollama

With Ollama installed, one command downloads and starts Phi-3.5 Mini 3.8B:

ollama run phi3.5:3.8b

Ollama's default download is Q4_K_M (about 6.1 GB here). You have room for FP16, which is full quality: pick that GGUF in LM Studio or from Hugging Face.

Run it in the cloud

No NVIDIA RTX 4090? Run Phi-3.5 Mini 3.8B in the cloud.

Rent a cloud GPU by the hour. No upfront cost, stop any time.

Memory needed by quantization

At 8,192 tokens of context. The black line marks the 24 GB the NVIDIA RTX 4090 can use.

QuantQualityMemoryFits?Speed
FP16Full quality11.5 GBFits~79 tok/s
Q8_0Near-lossless7.9 GBFits~149 tok/s
Q6_KExcellent7 GBFits~193 tok/s
Q5_K_MVery good6.5 GBFits~223 tok/s
Q4_K_MGood, most popular6.1 GBFits~261 tok/s
Q3_K_MNoticeably degraded5.7 GBFits~324 tok/s
Q2_KHeavily degraded5.3 GBFits~422 tok/s

Frequently asked questions

Can NVIDIA RTX 4090 run Phi-3.5 Mini 3.8B?

Yes. The NVIDIA RTX 4090 can run Phi-3.5 Mini 3.8B fully in VRAM at FP16 (full quality), using about 11.5 GB of the 24 GB available, at roughly 79 tokens/second.

How much VRAM does Phi-3.5 Mini 3.8B need?

About 6.1 GB at Q4_K_M with 8,192 tokens of context, and about 11.5 GB at full FP16 precision.

How fast is Phi-3.5 Mini 3.8B on the NVIDIA RTX 4090?

Roughly 79 tokens per second at FP16, estimated from the NVIDIA RTX 4090's 1008 GB/s memory bandwidth.