- Size
- 41.9B parameterscounted from the safetensors header on the Hub
- Licence
- custom, read it, MIT LicenseOur reading, check the licence: A custom licence we have not read for this model: other makers use the same name for different texts. Read the model's own licence file before you use it.Licence text
- On the Hub since
- 17 Aug 2024the date the repo was created
- Context length
- 128k tokensfrom the model's config
- Kind
- Base model (no base model declared on the Hub)PhiMoEForCausalLM
- On Hugging Face
- microsoft/Phi-3.5-MoE-instruct23k downloads in the last 30 days, 590 likes
Scores
Epoch's name: Phi-3.5-MoE-instruct. Best closed model on this set: Inflection-1, 89.7% (Inflection AI). All 136 results
Epoch's name: Phi-3.5-MoE-instruct. Best closed model on this set: gpt-4-0314, 92.0% (OpenAI). All 170 results
Epoch's name: Phi-3.5-MoE-instruct. Best closed model on this set: gpt-4o-mini-2024-07-18, 88.7% (OpenAI). All 113 results
Claimed by the maker Claimed
The model card has no claimed scores in a machine-readable form (the card's model-index), so none are shown. The maker may give scores in the card text: read the card.
Other views, which we link to and do not copy: look this model up on Artificial Analysis and the Arena leaderboard.
Independent scores from Epoch AI (CC-BY 4.0), fetched 19 Nov 2026: Epoch AI, ‘Capabilities & benchmarking’. Published online at epoch.ai. Retrieved from ‘https://epoch.ai/benchmarks’ [online resource]. Each row links its own source. What the badges mean.
Run it yourself
Memory needed = weights + context cache + 0.50 GB for the runtime. Context cache at 8k tokens: 1.0 GB, worked out from the model's config (microsoft/Phi-3.5-MoE-instruct), at 16-bit precision. Choose a machine above to see a verdict for each format. Method
GGUF for llama.cpp, Ollama, LM Studio and others
bartowski/Phi-3.5-MoE-instruct-GGUF (converted by bartowski), 2.8k downloads. Sizes are the real files on the Hub, all parts added up.
| Format | Size | Memory needed | Get it |
|---|---|---|---|
| IQ3_XS | 15.9 GBfile size | 17.4 GB15.9 + 1.0 + 0.50 | Commandsllama.cpp llama-server -hf bartowski/Phi-3.5-MoE-instruct-GGUF:IQ3_XSOllama ollama run hf.co/bartowski/Phi-3.5-MoE-instruct-GGUF:IQ3_XSDownload hf download bartowski/Phi-3.5-MoE-instruct-GGUF --include "Phi-3.5-MoE-instruct-IQ3_XS.gguf" --local-dir Phi-3.5-MoE-instruct-GGUF |
| Q3_K_S | 16.8 GBfile size | 18.3 GB16.8 + 1.0 + 0.50 | Commandsllama.cpp llama-server -hf bartowski/Phi-3.5-MoE-instruct-GGUF:Q3_K_SOllama ollama run hf.co/bartowski/Phi-3.5-MoE-instruct-GGUF:Q3_K_SDownload hf download bartowski/Phi-3.5-MoE-instruct-GGUF --include "Phi-3.5-MoE-instruct-Q3_K_S.gguf" --local-dir Phi-3.5-MoE-instruct-GGUF |
| IQ3_M | 17.1 GBfile size | 18.6 GB17.1 + 1.0 + 0.50 | Commandsllama.cpp llama-server -hf bartowski/Phi-3.5-MoE-instruct-GGUF:IQ3_MOllama ollama run hf.co/bartowski/Phi-3.5-MoE-instruct-GGUF:IQ3_MDownload hf download bartowski/Phi-3.5-MoE-instruct-GGUF --include "Phi-3.5-MoE-instruct-IQ3_M.gguf" --local-dir Phi-3.5-MoE-instruct-GGUF |
| Q3_K_M | 18.7 GBfile size | 20.2 GB18.7 + 1.0 + 0.50 | Commandsllama.cpp llama-server -hf bartowski/Phi-3.5-MoE-instruct-GGUF:Q3_K_MOllama ollama run hf.co/bartowski/Phi-3.5-MoE-instruct-GGUF:Q3_K_MDownload hf download bartowski/Phi-3.5-MoE-instruct-GGUF --include "Phi-3.5-MoE-instruct-Q3_K_M.gguf" --local-dir Phi-3.5-MoE-instruct-GGUF |
| Q3_K_L | 20.2 GBfile size | 21.7 GB20.2 + 1.0 + 0.50 | Commandsllama.cpp llama-server -hf bartowski/Phi-3.5-MoE-instruct-GGUF:Q3_K_LOllama ollama run hf.co/bartowski/Phi-3.5-MoE-instruct-GGUF:Q3_K_LDownload hf download bartowski/Phi-3.5-MoE-instruct-GGUF --include "Phi-3.5-MoE-instruct-Q3_K_L.gguf" --local-dir Phi-3.5-MoE-instruct-GGUF |
| Q3_K_XL | 20.3 GBfile size | 21.8 GB20.3 + 1.0 + 0.50 | Commandsllama.cpp llama-server -hf bartowski/Phi-3.5-MoE-instruct-GGUF:Q3_K_XLOllama ollama run hf.co/bartowski/Phi-3.5-MoE-instruct-GGUF:Q3_K_XLDownload hf download bartowski/Phi-3.5-MoE-instruct-GGUF --include "Phi-3.5-MoE-instruct-Q3_K_XL.gguf" --local-dir Phi-3.5-MoE-instruct-GGUF |
| IQ4_XS | 20.8 GBfile size | 22.3 GB20.8 + 1.0 + 0.50 | Commandsllama.cpp llama-server -hf bartowski/Phi-3.5-MoE-instruct-GGUF:IQ4_XSOllama ollama run hf.co/bartowski/Phi-3.5-MoE-instruct-GGUF:IQ4_XSDownload hf download bartowski/Phi-3.5-MoE-instruct-GGUF --include "Phi-3.5-MoE-instruct-IQ4_XS.gguf" --local-dir Phi-3.5-MoE-instruct-GGUF |
| IQ4_NL | 22.0 GBfile size | 23.5 GB22.0 + 1.0 + 0.50 | Commandsllama.cpp llama-server -hf bartowski/Phi-3.5-MoE-instruct-GGUF:IQ4_NLOllama ollama run hf.co/bartowski/Phi-3.5-MoE-instruct-GGUF:IQ4_NLDownload hf download bartowski/Phi-3.5-MoE-instruct-GGUF --include "Phi-3.5-MoE-instruct-IQ4_NL.gguf" --local-dir Phi-3.5-MoE-instruct-GGUF |
| Q4_0 | 22.1 GBfile size | 23.6 GB22.1 + 1.0 + 0.50 | Commandsllama.cpp llama-server -hf bartowski/Phi-3.5-MoE-instruct-GGUF:Q4_0Ollama ollama run hf.co/bartowski/Phi-3.5-MoE-instruct-GGUF:Q4_0Download hf download bartowski/Phi-3.5-MoE-instruct-GGUF --include "Phi-3.5-MoE-instruct-Q4_0.gguf" --local-dir Phi-3.5-MoE-instruct-GGUF |
| Q4_K_S | 22.2 GBfile size | 23.7 GB22.2 + 1.0 + 0.50 | Commandsllama.cpp llama-server -hf bartowski/Phi-3.5-MoE-instruct-GGUF:Q4_K_SOllama ollama run hf.co/bartowski/Phi-3.5-MoE-instruct-GGUF:Q4_K_SDownload hf download bartowski/Phi-3.5-MoE-instruct-GGUF --include "Phi-3.5-MoE-instruct-Q4_K_S.gguf" --local-dir Phi-3.5-MoE-instruct-GGUF |
| Q4_K_M | 23.6 GBfile size | 25.1 GB23.6 + 1.0 + 0.50 | Commandsllama.cpp llama-server -hf bartowski/Phi-3.5-MoE-instruct-GGUF:Q4_K_MOllama ollama run hf.co/bartowski/Phi-3.5-MoE-instruct-GGUF:Q4_K_MDownload hf download bartowski/Phi-3.5-MoE-instruct-GGUF --include "Phi-3.5-MoE-instruct-Q4_K_M.gguf" --local-dir Phi-3.5-MoE-instruct-GGUF |
| Q4_K_L | 23.7 GBfile size | 25.2 GB23.7 + 1.0 + 0.50 | Commandsllama.cpp llama-server -hf bartowski/Phi-3.5-MoE-instruct-GGUF:Q4_K_LOllama ollama run hf.co/bartowski/Phi-3.5-MoE-instruct-GGUF:Q4_K_LDownload hf download bartowski/Phi-3.5-MoE-instruct-GGUF --include "Phi-3.5-MoE-instruct-Q4_K_L.gguf" --local-dir Phi-3.5-MoE-instruct-GGUF |
| Q4_1 | 24.4 GBfile size | 25.9 GB24.4 + 1.0 + 0.50 | Commandsllama.cpp llama-server -hf bartowski/Phi-3.5-MoE-instruct-GGUF:Q4_1Ollama ollama run hf.co/bartowski/Phi-3.5-MoE-instruct-GGUF:Q4_1Download hf download bartowski/Phi-3.5-MoE-instruct-GGUF --include "Phi-3.5-MoE-instruct-Q4_1.gguf" --local-dir Phi-3.5-MoE-instruct-GGUF |
| Q5_K_S | 26.8 GBfile size | 28.3 GB26.8 + 1.0 + 0.50 | Commandsllama.cpp llama-server -hf bartowski/Phi-3.5-MoE-instruct-GGUF:Q5_K_SOllama ollama run hf.co/bartowski/Phi-3.5-MoE-instruct-GGUF:Q5_K_SDownload hf download bartowski/Phi-3.5-MoE-instruct-GGUF --include "Phi-3.5-MoE-instruct-Q5_K_S.gguf" --local-dir Phi-3.5-MoE-instruct-GGUF |
| Q5_K_M | 27.7 GBfile size | 29.2 GB27.7 + 1.0 + 0.50 | Commandsllama.cpp llama-server -hf bartowski/Phi-3.5-MoE-instruct-GGUF:Q5_K_MOllama ollama run hf.co/bartowski/Phi-3.5-MoE-instruct-GGUF:Q5_K_MDownload hf download bartowski/Phi-3.5-MoE-instruct-GGUF --include "Phi-3.5-MoE-instruct-Q5_K_M.gguf" --local-dir Phi-3.5-MoE-instruct-GGUF |
| Q5_K_L | 27.8 GBfile size | 29.3 GB27.8 + 1.0 + 0.50 | Commandsllama.cpp llama-server -hf bartowski/Phi-3.5-MoE-instruct-GGUF:Q5_K_LOllama ollama run hf.co/bartowski/Phi-3.5-MoE-instruct-GGUF:Q5_K_LDownload hf download bartowski/Phi-3.5-MoE-instruct-GGUF --include "Phi-3.5-MoE-instruct-Q5_K_L.gguf" --local-dir Phi-3.5-MoE-instruct-GGUF |
| Q6_K | 32.0 GBfile size | 33.5 GB32.0 + 1.0 + 0.50 | Commandsllama.cpp llama-server -hf bartowski/Phi-3.5-MoE-instruct-GGUF:Q6_KOllama ollama run hf.co/bartowski/Phi-3.5-MoE-instruct-GGUF:Q6_KDownload hf download bartowski/Phi-3.5-MoE-instruct-GGUF --include "Phi-3.5-MoE-instruct-Q6_K.gguf" --local-dir Phi-3.5-MoE-instruct-GGUF |
| Q6_K_L | 32.1 GBfile size | 33.6 GB32.1 + 1.0 + 0.50 | Commandsllama.cpp llama-server -hf bartowski/Phi-3.5-MoE-instruct-GGUF:Q6_K_LOllama ollama run hf.co/bartowski/Phi-3.5-MoE-instruct-GGUF:Q6_K_LDownload hf download bartowski/Phi-3.5-MoE-instruct-GGUF --include "Phi-3.5-MoE-instruct-Q6_K_L.gguf" --local-dir Phi-3.5-MoE-instruct-GGUF |
| Q8_0 | 41.4 GBfile size | 42.9 GB41.4 + 1.0 + 0.50 | Commandsllama.cpp llama-server -hf bartowski/Phi-3.5-MoE-instruct-GGUF:Q8_0Ollama ollama run hf.co/bartowski/Phi-3.5-MoE-instruct-GGUF:Q8_0Download hf download bartowski/Phi-3.5-MoE-instruct-GGUF --include "Phi-3.5-MoE-instruct-Q8_0.gguf" --local-dir Phi-3.5-MoE-instruct-GGUF |
| F16 | 78.0 GBfile size, 3 parts | 79.5 GB78.0 + 1.0 + 0.50 | Commandsllama.cpp llama-server -hf bartowski/Phi-3.5-MoE-instruct-GGUF:F16Download hf download bartowski/Phi-3.5-MoE-instruct-GGUF --include "Phi-3.5-MoE-instruct-f16/Phi-3.5-MoE-instruct-f16-*-of-00003.gguf" --local-dir Phi-3.5-MoE-instruct-GGUF |
Original weights size from the safetensors header, for vLLM, SGLang, Transformers
| Format | Size | Memory needed | Repo |
|---|---|---|---|
| BF16original weights | 78.0 GBfile size | 79.5 GB78.0 + 1.0 + 0.50 |
hf download microsoft/Phi-3.5-MoE-instruct --local-dir Phi-3.5-MoE-instructOther conversions (1; sizes not read)
- Kris2017/Phi-3.5-MoE-instruct-W4A16-AutoRound Other, w4a16, 28 downloads
The Hub tags 17 quantised repos as conversions of this model; we read the most downloaded and show the maker's own first.
Run it by API
No provider on Hugging Face Inference Providers serves this model (as of 19 Nov 2026, 13:58 UTC). Other hosts may; we only list what the router reports.
Not built yet: one key for every open model, with BlockRun, and our own tested-quality score per provider. Until then, precision and measured speed are the best guide to a provider here.
Planned for this page Not built yet
- Try it in private chat
- A link to Solana AI's confidential chat for the models it serves. This model is not listed as served there.
- Fine-tune it
- Fine-tuning a few base models from this page and publishing the result here.
- Creators and earnings
- Pages for people who publish fine-tunes and adapters of this model, and what they earn.
- Fund and bounties
- Bounties for new open models, paid from a public fund. How the fund will work.
- Contribute compute
- How to give compute to open model work, with Akash.
Model data from the Hugging Face Hub 19 Nov 2026, 13:43 UTC. The repo. Found a mistake on this page? Tell us.