LLM Perks

Phi 3.5 MoE

Microsoft · All Phi models
Free right now Mid-size
Open weights

Phi 3.5 MoE is a language model with 42B parameters from Microsoft. It is currently free at 1 provider with up to — tokens of context. Its weights are open, so it can also run locally.

Developer
Microsoft
Parameters
42B / 6.6B active
Type
Text
License
Open weights

Where to use Phi 3.5 MoE for free

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

ProviderContextLimitCard
NVIDIA Build
microsoft/phi-3.5-moe-instruct
— Free serverless inference for development; limits and model availability may change. No Open

Quick start

curl https://integrate.api.nvidia.com/v1/chat/completions \
  -H "Authorization: Bearer $NVIDIA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "microsoft/phi-3.5-moe-instruct",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Example for NVIDIA Build. 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 We do not publish our own estimates.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 3.5 MoE free?

Yes. As of September 26, 2026, Phi 3.5 MoE is free at NVIDIA Build. Each provider has its own limits — see the table above.

How do I call Phi 3.5 MoE via API?

Use any OpenAI-compatible client with base URL https://integrate.api.nvidia.com/v1, model ID microsoft/phi-3.5-moe-instruct, and a NVIDIA Build key.

Is Phi 3.5 MoE open-weight?

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