LLMs ·

Phi-4 Reasoning Vision 15B

Parameters
15B
Architecture
Dense
Max context
32K tokens
License
mit
Released
2026-01

How to run it

Original weights · Hugging Face

Full-precision files from Microsoft, for fine-tuning, vLLM, Transformers.

huggingface.co/microsoft/Phi-4-reasoning-vision-15B (opens in a new tab)

No sign-in needed.

Links checked 3 days ago against Hugging Face, Ollama and GitHub.

Apps that run Phi-4 Reasoning Vision 15B

Our pick for each kind of computer, assuming the model fits in its memory. Pick your own GPU or device in the Model Finder for an exact answer.

Mac (Apple silicon)

Docker Model Runner

Handy if you already work in Docker.

Also good:

Windows PC, NVIDIA GPU

Docker Model Runner

Handy if you already work in Docker.

Also good:

Windows PC, AMD Radeon

Docker Model Runner

Handy if you already work in Docker.

Also good:

Linux, NVIDIA GPU

Docker Model Runner

Handy if you already work in Docker.

Also good:

Linux server, 80 GB+ GPUs

vLLM

The full model fits in GPU memory at 8-bit, so it can batch many requests: best for serving a team or an app.

Also good:

App Type Opens it as Systems Engine Start
vLLM Server Original weights WSLWindowsLinux vLLM 
vllm serve microsoft/Phi-4-reasoning-vision-15B
SGLang Server Original weights WSLWindowsLinux SGLang 
python3 -m sglang.launch_server --model-path microsoft/Phi-4-reasoning-vision-15B
Docker Model Runner CLI + API Original weights macOSWindowsLinux llama.cpp, vLLM 
docker model run hf.co/microsoft/Phi-4-reasoning-vision-15B
Text Generation WebUI Desktop app Original weights macOSWindowsLinux llama.cpp, Transformers, ExLlamaV3 Get Text Generation WebUI

Which apps can open a model depends on the files published for it: GGUF for llama.cpp-based apps, MLX for Apple’s engine, the original weights for vLLM and SGLang. Compare every app.

Memory needed

Weights plus KV cache and runtime overhead. Leave a little extra free for the operating system and other apps. Try another context length.

Quantization Quality Weights Total at 8K context Total at 32K context Download
Q2_K Low 6.0 GB 8.3 GB 13.4 GB All files (opens in a new tab)
Q3_K_M Low 7.4 GB 9.7 GB 14.7 GB All files (opens in a new tab)
Q4_K_M Good 9.2 GB 11.5 GB 16.5 GB All files (opens in a new tab)
Q5_K_M High 10.7 GB 13.0 GB 18.0 GB All files (opens in a new tab)
Q6_K High 12.4 GB 14.7 GB 19.7 GB All files (opens in a new tab)
Q8_0 Very high 16.0 GB 18.3 GB 23.3 GB All files (opens in a new tab)
F16 Full 30.2 GB 32.5 GB 37.6 GB Original weights (opens in a new tab)

Best hardware for Phi-4 Reasoning Vision 15B

The least expensive GPU in our catalog that runs it without offloading is the Intel Arc B570 (launch price $219) at Q2_K.

Quality Hardware Quant Memory Speed Runs
70Great quality (A) · 70 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 4090 Founders EditionNVIDIA GeForce RTX 4090 Q8_0 19.5 GB / 24.0 GB 41.7 tok/s Runs great
70Great quality (A) · 70 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. ASUS ROG Strix GeForce RTX 3090 OC Edition 24GB GDDR6XNVIDIA GeForce RTX 3090 Q8_0 19.5 GB / 24.0 GB 38.7 tok/s Runs great
69Good quality (B) · 69 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. Sapphire NITRO+ AMD Radeon RX 7900 XTX Vapor-X 24GBAMD Radeon RX 7900 XTX Q6_K 15.9 GB / 24.0 GB 46.6 tok/s Runs great
70Great quality (A) · 70 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 5090 Founders EditionNVIDIA GeForce RTX 5090 Q8_0 19.9 GB / 32.0 GB 74.1 tok/s Runs great
69Good quality (B) · 69 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 4080 SUPERNVIDIA GeForce RTX 4080 SUPER Q6_K 15.5 GB / 16.0 GB 41.5 tok/s Runs welltight fit
69Good quality (B) · 69 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. ASUS TUF Gaming GeForce RTX 5070 Ti 16GB GDDR7 OC EditionNVIDIA GeForce RTX 5070 Ti Q6_K 15.5 GB / 16.0 GB 50.5 tok/s Runs welltight fit
69Good quality (B) · 69 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 5080 Founders EditionNVIDIA GeForce RTX 5080 Q6_K 15.5 GB / 16.0 GB 54.1 tok/s Runs welltight fit
66Good quality (B) · 66 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. Sapphire NITRO+ AMD Radeon RX 9070 XT OCAMD Radeon RX 9070 XT Q4_K_M 12.3 GB / 16.0 GB 44.7 tok/s Runs great
66Good quality (B) · 66 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 4070 Ti SUPERNVIDIA GeForce RTX 4070 Ti SUPER Q4_K_M 12.3 GB / 16.0 GB 54.6 tok/s Runs great
66Good quality (B) · 66 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. ASUS Dual GeForce RTX 5060 Ti 16GB GDDR7 OC EditionNVIDIA GeForce RTX 5060 Ti 16GB Q4_K_M 12.3 GB / 16.0 GB 36.4 tok/s Runs great

Check it on your own hardware

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.

Was this page helpful?