Wednesday, October 28, 2026 🏢 AI Companies Hub RSS About Contact Admin
POPULAR BEATS: Generative AI LLMs & NLP Autonomous Agents Robotics & Hardware Enterprise AI AI Ethics & Policy 🏢 All AI Companies

Phi-4 14B

Synthetic Data Small Language Model (SLM)
Official Source & Paper
Microsoft AI logo Microsoft 📄 Open Weights (MIT) 🌍 United States (US) 🗓️ Effective Date: 12.12.2024

Microsoft's breakthrough 14B model trained heavily on curated synthetic textbooks, outperforming models five times its size on mathematical reasoning.

Model Size / Parameters 14B Dense Architecture: Synthetic Data Small Language Model (SLM)
Context Window (Input) 16k tokens Token Capacity
Max Output Tokens 4,096 tokens Single Generation Limit
Supported Modalities Text, Code Native Multimodal Ingestion
License & Access Open Weights (MIT) Open Weights vs. API
Release Date December 2024 Effective Date: 12.12.2024

Verified Benchmark Results

Standardized academic and industry evaluation benchmarks (Effective Date: 12.12.2024).

MMLU-Pro / 5-Shot General Intellect
84.8%
SWE-bench Verified Software Engineering
39.4%
MATH 500 / AIME Mathematical Reasoning
80.4%
GPQA Diamond Doctoral Science
44.8%
HumanEval / Python Algorithm Synthesis
84.1%

Pricing & Cost Efficiency

Official API token pricing and self-hosted licensing terms.

Input Pricing (Prompt)
Free / Self-Hosted
Per 1 million processed tokens
Output Pricing (Generation)
Free / Self-Hosted
Per 1 million generated tokens
Verified Datasheet: Source: Microsoft Research Documentation • Effective Date: 12.12.2024
← Back to All AI Model Datasheets

Latest Coverage & Analysis on Phi-4 14B