alt.freeweights.phi / Phi-3-mini-4k-instruct
From: unsloth <huggingface.co/unsloth>Newsgroups: alt.freeweights.phiSubject:

Phi-3-mini-4k-instruct

Date: 2026-11-15Commit: 636c707Signed: 2DC28CC7B26A283B ✓

Text generation model with 3.8B parameters. Safetensors weights, 7.64 GB in all. Published by unsloth under MIT.

3.8B params7.64GBPhi-3-mini-4k-instructsafetensorspermissive
FactsParams3.8BSize7.64 GBFormatssafetensorsTasktext generationLicenseMITFrom sourcepermissive
$ fw get unsloth/Phi-3-mini-4k-instruct
# manifest sha256 7a087181a2ba… · 7.64 GB · 12 files
$ fw verify ./Phi-3-mini-4k-instruct unsloth/Phi-3-mini-4k-instruct
From sourceHosted (signed)unsloth/Phi-3-mini-4k-instructLast check: 2026-11-15 03:30 UTC

Get help: custom fine-tune or private deploy. We review your setup before we quote.

Works with

Overview

Size
7.64 GB
Weights
7.64 GB (safetensors)
Parameters
3.8B
Formats
safetensors
Task
text generation
License
MIT (permissive)
Files
12, each with a sha256
Publisher
unsloth
Added
2026-11-15
Section
alt.freeweights.phi
Lineage

Original release from the publisher.

Where these bytes come from
Publisher at the pinned commit→You

Sources are tried in order. A file counts only when its size and sha256 match the manifest. Sources are checked when you ask for them, not on a timer.

Sources
BadgeHostLast checkResult
From sourcehuggingface.co2026-11-15 03:30 UTCpassed

Model card

From the publisher at commit 636c707. Shown as plain text with links; images open at the publisher.

Reminder to use the dev version Transformers:#

pip install git+https://github.com/huggingface/transformers.git

Finetune Phi-3, Llama 3, Gemma 2, Mistral 2-5x faster with 70% less memory via Unsloth!#

Directly quantized 4bit model with bitsandbytes. We Mistralfied the model to ensure it could be used on many platforms

We have a Google Colab Tesla T4 notebook for Phi-3 (mini) here: https://colab.research.google.com/drive/1vIrqH5uYDQwsJ4-OO3DErvuv4pBgVwk4?usp=sharing And another notebook for Phi-3 (medium) here: https://colab.research.google.com/drive/1hhdhBa1j_hsymiW9m-WzxQtgqTH_NHqi?usp=sharing

https://discord.gg/u54VK8m8tk https://ko-fi.com/unsloth https://github.com/unslothai/unsloth

✨ Finetune for Free#

All notebooks are beginner friendly! Add your dataset, click "Run All", and you'll get a 2x faster finetuned model which can be exported to GGUF, vLLM or uploaded to Hugging Face.

Unsloth supportsFree NotebooksPerformanceMemory use
Llama 3 (8B)▶️ Start on Colab2.4x faster58% less
Gemma 2 (9B)▶️ Start on Colab2x faster63% less
Mistral (9B)▶️ Start on Colab2.2x faster62% less
Phi 3 (mini)▶️ Start on Colab2x faster50% less
Phi 3 (medium)▶️ Start on Colab2x faster50% less
TinyLlama▶️ Start on Colab3.9x faster74% less
CodeLlama (34B) A100▶️ Start on Colab1.9x faster27% less
Mistral (7B) 1xT4▶️ Start on Kaggle5x faster\*62% less
DPO - Zephyr▶️ Start on Colab1.9x faster19% less

Files

12 files · 7.64 GBWeights 7.64 GB (safetensors). Sizes and sha256 from the signed manifest.
Hide all 12 files with their sha256
FileSizesha256
.gitattributes1.52 KB11ad7efa24975ee4b0c3c3a38ed18737f0658a5f75a0a96787b576a78a023361
README.md3.81 KB86ac560257b5dcc2fceb6a0d792b46ee01f8631bfb11591a8c7fe86195902713
added_tokens.json293 B5f5c94d30901fe9c961d776d51eadbda4454991a77b53d36e0be0571a8e72a7d
config.json724 Ba301041e0e0721cc86f66164716807e33fecd06f0ea4066e3ef8fb5d3905973b
generation_config.json194 B1fcfb7e28d220a400e118f5f84066e69b46c56804f1e48f6fb6fdb17bbe2d4e0
model-00001-of-00002.safetensors4.99 GB3403fc7e50e4a23a265a54ee7c74c465be32dcf93371d5346d38b9a724f0546f
model-00002-of-00002.safetensors2.65 GB79e3d00e5431b1b409d0222aea0daa244f715dc8a030b0dbc5e44be9b7353112
model.safetensors.index.json23.9 KBc61cc0451bf0e12a51c3ee0ae364aa1bf2a744bf6136080dc297455a737e9575
special_tokens_map.json458 B1a72bd7fd5ba0070fe8303e89099b0b667436dce3b67dc3405c671783964b2f2
tokenizer.json1.84 MBccbb651863e53355593cf1227c115a02d215a579fef875f04f6728fc81bdfa56
tokenizer.model500 KB9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
tokenizer_config.json3.34 KB2183895a58f36971081c0d952f4286cd5a28e210cf5e20b16abdc9621cd6c6d4
VersionsEach time the pinned upstream commit or file list changed in this repo. Newest first.
Upstream commitDateContents
636c707430 pinned now2026-11-15 03:30 UTC12 files · 7.64 GB

Use it

$ fw get unsloth/Phi-3-mini-4k-instruct
# manifest sha256 7a087181a2ba… · 7.64 GB · 12 files
$ fw verify ./Phi-3-mini-4k-instruct unsloth/Phi-3-mini-4k-instruct
fw.pyChecks the signature and every sha256, tries each source in order, resumes.
$ python3 fw.py get https://freeweights.allternit.com/manifests/unsloth/Phi-3-mini-4k-instruct.json --dir ./Phi-3-mini-4k-instruct $ python3 fw.py verify ./Phi-3-mini-4k-instruct ./unsloth__Phi-3-mini-4k-instruct.json
hfThe Hugging Face CLI with one setting. Same repo id.
$ HF_ENDPOINT=https://freeweights.allternit.com hf download unsloth/Phi-3-mini-4k-instruct --revision 636c707430a5509c80b1aa51d05c127ed339a975 --local-dir ./Phi-3-mini-4k-instruct
transformersLoad the weights in Python.
import os os.environ["HF_ENDPOINT"] = "https://freeweights.allternit.com" from transformers import AutoModelForCausalLM, AutoTokenizer tok = AutoTokenizer.from_pretrained("unsloth/Phi-3-mini-4k-instruct", revision="636c707430a5509c80b1aa51d05c127ed339a975") model = AutoModelForCausalLM.from_pretrained("unsloth/Phi-3-mini-4k-instruct", revision="636c707430a5509c80b1aa51d05c127ed339a975", device_map="auto")
vLLMServe an OpenAI-compatible API.
$ HF_ENDPOINT=https://freeweights.allternit.com vllm serve unsloth/Phi-3-mini-4k-instruct --revision 636c707430a5509c80b1aa51d05c127ed339a975

License

Name
MIT License
Id
MIT
Class
Permissive
Redistribution
yes
Listing
Hosted (signed)
Decided
License as declared by the publisher. No person read it. 2026-11-15
Text in
README.md
Text sha256
86ac560257b5dcc2fceb6a0d792b46ee01f8631bfb11591a8c7fe86195902713
Terms gate
No extra terms beyond the license.

Read the license text

Licenses are checked against the license table. A custom license passes only after a person reads it.

Provenance

From: unsloth <huggingface.co/unsloth>Newsgroups: alt.freeweights.phiSubject: Phi-3-mini-4k-instructDate: 2026-11-15Commit: 636c707Signed: 2DC28CC7B26A283B ✓
Upstream
unsloth/Phi-3-mini-4k-instruct
Pinned commit
636c707430a5509c80b1aa51d05c127ed339a975
Manifest
Phi-3-mini-4k-instruct.json · .minisig
Manifest sha256
7a087181a2bac923c4cd97232bae86ff36b6472c23b24cd8f06aff516a856260
Signature
minisign, Ed25519, key 2DC28CC7B26A283B
Created
2026-11-15 03:30 UTC
Generator
freeweights-manifest/0.1.0
Takedowns
None.

How to verify

$ curl -O https://freeweights.allternit.com/manifests/unsloth/Phi-3-mini-4k-instruct.json $ curl -O https://freeweights.allternit.com/manifests/unsloth/Phi-3-mini-4k-instruct.json.minisig $ minisign -Vm Phi-3-mini-4k-instruct.json -P RWQ7KGqyx4zCLXtUFOqvfv8L3g+8fpKJ1FsuxPyOYMF+4KufbUxKq1Oq $ shasum -a 256 <file> # compare with the Files tab

A manifest whose signature fails must be rejected. A file whose size or sha256 does not match must be discarded. Verify a download

Model tree

No related models in this catalog.Nothing here lists this model as its parent, and it names no parent.
Get it
Download
$ HF_ENDPOINT=https://freeweights.allternit.com hf download unsloth/Phi-3-mini-4k-instruct
Open in Freeweights Studio

Pull, verify and run it in the Allternit app on Desktop, web or phone.

Signed manifest
minisign, key 2DC28CC7B26A283B. Phi-3-mini-4k-instruct.json. Provenance. Takedown history: none.
From: BAAI <huggingface.co/BAAI>Subject: bunny-phi-2-siglip-lora
From: microsoft <huggingface.co/microsoft>Subject: MediPhi
From: microsoft <huggingface.co/microsoft>Subject: MediPhi-Clinical
From: microsoft <huggingface.co/microsoft>Subject: MediPhi-Guidelines
From: microsoft <huggingface.co/microsoft>Subject: MediPhi-Instruct
From: microsoft <huggingface.co/microsoft>Subject: MediPhi-MedCode