New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
Phi-3.5-MoE
About Phi-3.5-MoE
Phi-3.5-MoE is an AI model developed by Microsoft, in the language category, released in 2024, made available as an open-weights model with 60.8B params.
On this page you'll find Phi-3.5-MoE's full specifications, including 16 benchmark results. Review provider pricing and benchmark scores below, or compare Phi-3.5-MoE head-to-head with other language models.
Links
Run Phi-3.5-MoE locally — what to buy
Share this build ↗Sized to what Phi-3.5-MoE actually needs (~8k context at Q4), not to the biggest GPU: Good = cheapest that runs it, Better = best value, Best = most headroom. Speed is a hardware estimate; anything past ~120 tok/s is shown as instant.
Software support
✓ measured · · compatible · — not supported. Informational only — speed is hardware-based.
Community
No reports yet. Own this model on some hardware? Be the first to confirm it runs.
No comments yet. Start the discussion.
Phi-3.5-MoE benchmark scores▶
| Benchmark | Category | Score | Variant |
|---|---|---|---|
| HumanEval | code | 70.7% | instruct-0-shot |
| MBPP | code | 80.8% | instruct-3-shot |
| ArenaHard | general | 37.9% | instruct-from-hf-readme |
| MMLU | general | 78.9% | instruct-5-shot |
| MMLU-Pro | general | 54.3% | instruct-0-shot-cot |
| MMMLU | general | 69.9% | instruct-multilingual-5-shot |
| ARC-Challenge | knowledge | 91 | instruct-10-shot |
| GSM8K | math | 88.7% | instruct-8-shot-cot |
| MATH | math | 59.5% | instruct-0-shot-cot |
| MGSM | math | 58.7 | instruct-0-shot-cot |
| BIG-Bench Hard | reasoning | 79.1% | instruct-0-shot-cot |
| GPQA Diamond | reasoning | 36.8% | instruct-0-shot-cot |
| HellaSwag | reasoning | 83.8 | instruct-5-shot |
| PIQA | reasoning | 88.6 | instruct-5-shot |
| Winogrande | reasoning | 81.3 | instruct-5-shot |
| TruthfulQA | safety | 77.5% | instruct-mc2-10-shot |
More models like Phi-3.5-MoE
Frequently asked questions
What is Phi-3.5-MoE? ▶
Phi-3.5-MoE is an AI model developed by Microsoft, in the language category, released in 2024. It is tracked on GenAIList with its specifications, benchmark scores and provider pricing.
Who created Phi-3.5-MoE? ▶
Phi-3.5-MoE was developed by Microsoft and released in 2024.
Is Phi-3.5-MoE open source or proprietary? ▶
Phi-3.5-MoE ships as an open-weights model: you can download and self-host the weights, though the licence may place some restrictions on use.
How much does Phi-3.5-MoE cost? ▶
Pricing for Phi-3.5-MoE depends on the provider. See the providers table on this page for the latest API rates.
How does Phi-3.5-MoE perform on benchmarks? ▶
Phi-3.5-MoE is benchmarked across 16 evaluations on GenAIList, including HumanEval (70.7%). See the full benchmark table below and compare it with other models.
No videos yet. Know a good one? Add it below.
Know a great video about Phi-3.5-MoE?
Add YouTube Video
Add Tweet/Discussion
Report an Issue
Help us maintain accurate information in our database.
Suggest an Edit
Tell us what needs to be changed or corrected for