Phi-4 VRAM and GPU Requirements

Phi-4 needs about 10.24 GB at Q4_K_M with 8K context.

Which GPUs fit

VRAM by quantization

Totals at 8,192 tokens. Q4_K_M is the usual balance of size and quality. FP16 weights are the parameter count times 2 bytes. The FP16 column in the table is the full estimate: weights, cache, and 1 GB.

Context length

Q4_K_M total as the context changes. Cache is the part that grows.

• 2,048 9.07 GB
• 8,192 10.24 GB
• 16,384 11.80 GB

Phi-4 hardware notes

• Unified memory counts 75 percent of the listed pool as usable. Discrete cards use the full listed memory. The memory column still shows the listed figure.
• Last updated 2026-07-14.
• Weights, cache, and the 1 GB overhead are estimates. How this is calculated.
• Architecture numbers and the parameter count come from the public model record, checked 2026-10-10.

Phi-4 VRAM FAQ

How much VRAM does Phi-4 need at Q4_K_M?

10.24 GB at 8,192 tokens. Weights are 7.68 GB, the cache is 1.56 GB, and overhead is 1 GB.

Does a 24 GB GPU fit Phi-4 at Q4_K_M?

GeForce RTX 3090 24GB is Fits, 13.76 GB left. The quant table uses 8,192 tokens, the smaller of 8,192 and the configured maximum of 16,384.

What context length is configured for Phi-4?

The configured context length is 16,384 tokens. The quant table uses 8,192 tokens, the smaller of 8,192 and the configured maximum of 16,384.

Which source supplies the parameter count for Phi-4?

14.66B parameters, from the Hugging Face model record for microsoft/phi-4.

Related models