LLM Perks

Phi 3 Mini 4K

Microsoft · All Phi models
Free right now Small
Open weights

Phi 3 Mini 4K is a language model with 3.8B parameters from Microsoft. It is currently free at 1 provider with up to 4K tokens of context. Its weights are open, so it can also run locally.

Developer
Microsoft
Parameters
3.8B
Context
4K tokens
Type
Text
License
Open weights

Where to use Phi 3 Mini 4K for free

Each row is a separate free endpoint with its own limits.

ProviderContextLimitCard
Hugging Face
microsoft/Phi-3-mini-4k-instruct
4K Fair use (depends on server load) No Open

Quick start

curl https://router.huggingface.co/v1/chat/completions \
  -H "Authorization: Bearer $HUGGINGFACE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "microsoft/Phi-3-mini-4k-instruct",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Example for Hugging Face. Any OpenAI-compatible client works: Cursor, Cline, OpenCode, Open WebUI.

Benchmarks

This model is not on the Artificial Analysis leaderboard yet. We do not publish our own estimates.

Similar free models

FAQ

Is Phi 3 Mini 4K free?

Yes. As of September 26, 2026, Phi 3 Mini 4K is free at Hugging Face. Each provider has its own limits — see the table above.

What is the context window of Phi 3 Mini 4K?

The largest context among free routes is 4,096 tokens. Some providers cap it lower.

How do I call Phi 3 Mini 4K via API?

Use any OpenAI-compatible client with base URL https://router.huggingface.co/v1, model ID microsoft/Phi-3-mini-4k-instruct, and a Hugging Face key.

Is Phi 3 Mini 4K open-weight?

Yes, the weights are published — you can download and run it locally, for example with Ollama or LM Studio.