Skip to content
03 Model garden Serve

Phi 3.5 MoE instruct

Runs as your own dedicated GPU deployment, ready to launch from the wizard. Data stays in Europe.

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

Call it now

Drop-in replacement for OpenAI: change only the base URL and API key. The Anthropic format (/v1/messages) is supported too.

Serve Model garden vLLM ready
microsoft/Phi-3.5-MoE-instruct

text->text · microsoft · sovereign EU

Runs on EU infrastructure operated by European companies, from your request to the GPU. See the full chain

Phi-3.5-MoE is a lightweight, state-of-the-art open model built upon datasets used for Phi-3 - synthetic data and filtered publicly available documents - with a focus on very high-quality, reasoning dense data. The model supports multilingual and comes with 128K context length (i...

Router
Router · on request
Dedicated
Dedicated · available
42BParameters
131KContext window
160GBMinimum VRAM
POST /api/v1/chat/completions On request

Specifications

Parameters42B
Context window131,072 tokens
Minimum VRAM160 GB
ArchitecturePhiMoEForCausalLM (vLLM)
Licensemit
Modalitytext->text
ReleasedAugust 2024
Publishermicrosoft ↗

Pricing

Shared router · per token

On request Not available on the shared router. Pricing on request as a dedicated GPU deployment.

Dedicated GPU · per hour

from €6,60 per hourYour own vLLM instance on European cloud (160 GB VRAM), billed per minute.

Shared EU router, you pay only per token. Dedicated GPU deployments are billed per minute, see pricing.

✓ Verified working on 30-07-2026, responded in 33818 ms on our EU infrastructure.

Bench index

Bench is our quality index for every model, 0 to 100. It rests on public measurements (Epoch AI with about 40 benchmarks, LiveBench, LMArena and more) and is refreshed every morning. Where nobody measures, it shows an estimate with a range.

38.2 / 100 · Via LiveBench

Translated from its LiveBench result onto the same scale.

Range 31.1 to 45.4

Sources: Epoch AI ↗ · LiveBench ↗ · LMArena ↗ · Refreshed daily, last on 21-11-2026

LiveBench · 34.1

Reasoning38.7
Coding21.7
Mathematics26.8
Data analysis40.5
Language17.1
Instruction following59.7
LiveBench · All tasks
Reasoning
web of lies v248.0
zebra puzzle34.0
spatial34.0
Coding
LCB generation29.5
coding completion14.0
Mathematics
AMPS Hard39.0
math comp21.9
olympiad19.5
Data analysis
cta60.0
tablejoin3.4
tablereformat58.0
Language
connections11.0
plot unscrambling14.2
typos26.0
Instruction following
paraphrase66.3
simplify63.3
story generation57.8
summarize51.6
Questions

Frequently asked questions

Can I run Phi 3.5 MoE instruct in the EU?

Yes. HostYourAI runs Phi 3.5 MoE instruct on GPUs in European datacenters via vLLM. Prompts and outputs never leave the EU and there is no US cloud provider in the chain.

Is hosting Phi 3.5 MoE instruct GDPR-compliant?

Yes. All processing happens inside the EU, a Data Processing Agreement (DPA) is available and the subprocessor list is public. Open-source weights also mean: no training on your data.

How much does Phi 3.5 MoE instruct cost?

Phi 3.5 MoE instruct needs several GPUs at once, so it runs as a dedicated deployment billed per GPU minute rather than per token. Tell us your volume and we will work it out with you.

Is the API OpenAI-compatible?

Yes. You use the standard OpenAI SDKs with a custom base URL (https://hostyourai.com/api/v1). The Anthropic Messages API is supported as a drop-in as well.

More models from Microsoft

The platform

One platform, four layers

Start with one layer and add the others when the work asks for it. They share one key, one balance and one Activity view.

Code

Your coding agent on open models in Europe, set up with one command.

Open

Control

Keys, budgets and guardrails per team, and personal data stopped before any model sees it.

Open

Serve

Shared per token, or running for you alone, from European data centres.

Open

EU compute

Private GPUs in Europe, H100 to B300, sized for response time, generation speed and concurrent users.

Open

Try Phi 3.5 MoE instruct for free

Creating an account takes a minute. Test Phi 3.5 MoE instruct straight away in the playground.

  • Jantine
  • Hans
  • Sander
  • Jos
  • Daan
  • Kyra
  • Margriet
  • Stan
The team behind HostYourAI