When was Phi-3.5-MoE-instruct released?

Phi-3.5-MoE-instruct was released on August 23, 2024 by Microsoft. This is the official Phi-3.5-MoE-instruct release date tracked on LLM Stats.

Is Phi-3.5-MoE-instruct available via API?

Yes, Phi-3.5-MoE-instruct is available via API. See the official documentation for authentication and endpoint details.

How big is Phi-3.5-MoE-instruct?

Phi-3.5-MoE-instruct has 60 billion parameters. It was trained on 4.9 trillion tokens. It ships as an open-weight model, so you can download and run it on your own hardware.

Who created Phi-3.5-MoE-instruct?

Phi-3.5-MoE-instruct was created by Microsoft.

What is the license for Phi-3.5-MoE-instruct?

Phi-3.5-MoE-instruct is released under the MIT license. This is an open-source / open-weight license that permits self-hosting.

Where is the Phi-3.5-MoE-instruct paper or technical report?

Phi-3.5-MoE-instruct has a paper or technical report available at https://arxiv.org/abs/2404.14219. Use that source for architecture, training, release and evaluation details.

What models should I compare Phi-3.5-MoE-instruct against?

Common Phi-3.5-MoE-instruct comparisons include Phi-3.5-MoE-instruct vs Claude 3 Opus, Phi-3.5-MoE-instruct vs Claude 3 Sonnet, Phi-3.5-MoE-instruct vs Gemini 1.5 Flash. Compare them side by side for benchmark scores, pricing, context window, latency and API availability.

Phi-3.5-MoE-instruct API Pricing, Context Window & Benchmarks

Name: Phi-3.5-MoE-instruct
Author: Microsoft

Phi-3.5-MoE-instruct: API Pricing, Context Window & Benchmarks

Phi-3.5-MoE-instruct is a language model from Microsoft, released in August 2024.

Phi-3.5-MoE-instruct is a mixture-of-experts model with ~42B total parameters (6.6B active) and a 128K context window. It excels at reasoning, math, coding, and multilingual tasks, outperforming larger dense models in many benchmarks. It

Phi-3.5-MoE-instruct benchmarks

Capability tiers

Standing within each category, adjusted for leaderboard depth.

Real tasks performance

High-confidence performance for Phi-3.5-MoE-instruct across real-world prompt categories. Only 95% intervals at most 4 points wide are shown.

Performance by conversation depth

How Phi-3.5-MoE-instruct holds up as conversations get longer.

Quality Tracker

Phi-3.5-MoE-instruct Performance Across Datasets

Scores sourced from the model's scorecard, paper, or official blog posts

llm-stats.com - Mon Sep 07 2026

Notice missing or incorrect data?

Phi-3.5-MoE-instruct model size

Phi-3.5-MoE-instruct has 60 billion parameters and was trained on 4.9 trillion tokens. See how it compares to other models in the same parameter range.

ParametersTraining tokens

60B

4.9Ttokens

82× tokens-to-params ratio

Large (30–80B)

60B

1B7B70B405B

Phi-3.5-MoE-instruct API

Available from the model provider

Phi-3.5-MoE-instruct has an official provider API. It is not currently routed through the LLM Stats gateway.

Read the official API documentation

Phi-3.5-MoE-instruct latency

Phi-3.5-MoE-instruct time to first token, sustained output throughput, and failed-request rate from live API traffic over the trailing 7 days.

Phi-3.5-MoE-instruct examples

Recent arena outputs from Phi-3.5-MoE-instruct, picked from the highest-ranked matchups.

Phi-3.5-MoE-instruct license

Phi-3.5-MoE-instruct is released under the MIT license, which permits commercial use, has 60.0B parameters.

License: MIT; Commercial use allowed
Parameters: 60.0B

MIT License - allows commercial use

Phi-3.5-MoE-instruct resources

Official sources for Phi-3.5-MoE-instruct: api documentation, paper or system card, official launch post.

Phi-3.5-MoE-instruct vs other models

The most-compared alternatives to Phi-3.5-MoE-instruct are Claude 3 Opus, Claude 3 Sonnet, Gemini 1.5 Flash. Open any pair side-by-side for benchmarks, pricing, context, and latency.