microsoft / text-generation updated 3 weeks ago

phi-4

Name: phi-4
Author: microsoft

-------------------------------------------------------------------------------------------------------- Developers Microsoft Research Description phi-4 is a state-of-the-art open model built upon a blend of synthetic datasets, data from filtered public domain websites, and acquired academic books and Q&A datasets. The goal of this approach was to ensure that small capable models were trained with data focused on hig...

Params

14.7 B

Context

16,384

Downloads 30d

671 K

Likes

2,286

Commercial use: allowed mit Not gated SAFETENSORS 1 languages View on Hugging Face ↗

Download history

daily snapshots · 10 days

713 K655 K

Jul 28Jul 31Aug 3Aug 6

Can you run it?

Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.

File	Quant	Size	Est. VRAM	Verdict on RTX 4090 · 24 GB
model.safetensors	bf16	29.3 GB	34.9 GB	❌ Won’t fit
model.safetensors (bf16, full)	bf16 + 16K ctx	29.3 GB	37.1 GB	❌ Won’t fit

Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.

Run it

copy-paste, exact tags checked against the Hub

~ · curl · api/v1

$ curl -s https://aimodelscomparison.com/api/v1/models/phi-4
{
  "hf_id": "microsoft/phi-4",
  "params_b": 14.66,
  "context_length": 16384,
  "license": { "id": "mit", "commercial": "yes" },
  "downloads_30d": 670910,
  "vram_estimates": [
    { "quant": "bf16", "gb": 34.9 }
  ],
  "updated_at": "2026-07-28T18:04:14Z"
}

est. VRAM —on RTX 4090 · 24 GBJSON API →

Specifications

Architecture: Phi3ForCausalLM
Parameters: 14.7 B
Tensor type: BF16
Context length: 16,384
Vocabulary: 100,352
Layers / heads: 40 / 40
Licence: mit
First seen on the Hub: 2024-12-11
Training datasets: undisclosed
Added to our catalog: 2026-07-28

Compare with

Qwen3-8B 16.3 M Qwen2.5-7B-Instruct 12.1 M Qwen3.6-35B-A3B-NVFP4 11.3 M gpt-oss-20b 8.6 M Qwen3-32B 8.3 M Llama-3.1-8B-Instruct 8.0 M

Appears in

Best open coding models Commercial-use text models Models that fit in 12 GB