CatalogWantedMirrorsLedgerOperating StatementDocsStudio statusCard system
alt.freeweights.text / Jan-v3-4B-base-instruct
From: janhq <huggingface.co/janhq>Newsgroups: alt.freeweights.textSubject:

Jan-v3-4B-base-instruct

Date: 2026-11-15Commit: f331713Signed: 2DC28CC7B26A283B ✓

Text generation model with 4.4B parameters. Safetensors weights, 8.84 GB in all. Published by janhq under Apache-2.0.

4.4B params8.84GBJan-v3-4B-base-instructsafetensorspermissive
FactsParams4.4BSize8.84 GBFormatssafetensorsTasktext generationLicenseApache-2.0From sourcepermissive
$ fw get janhq/Jan-v3-4B-base-instruct
# manifest sha256 0af9580d8c1a… · 8.84 GB · 14 files
$ fw verify ./Jan-v3-4B-base-instruct janhq/Jan-v3-4B-base-instruct
From sourceHosted (signed)janhq/Jan-v3-4B-base-instructLast check: 2026-11-15 03:42 UTC
Run at homeGet help

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

Fine-tune of Qwen3-4B-Instruct-2507.

Works with

Run24Fine-tune29Convert15Evaluate13Scan13Apps58
OverviewModel cardFilesUse itLicenseProvenanceModel tree

Overview

Size
8.84 GB
Weights
8.82 GB (safetensors)
Parameters
4.4B
Formats
safetensors
Task
text generation
License
Apache-2.0 (permissive)
Files
14, each with a sha256
Publisher
janhq
Added
2026-11-15
Section
alt.freeweights.text
Lineage

Original release from the publisher.

Base model: Qwen/Qwen3-4B-Instruct-2507

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 03:42 UTCpassed

Model card

From the publisher at commit f331713. Shown as plain text with links; images open at the publisher.

Jan-v3-4B-base-instruct: a 4B baseline model for fine-tuning#

GitHub License Jan App

Image: image

Overview#

Jan-v3-4B-base-instruct is a 4B-parameter model obtained via post-training distillation from a larger teacher, transferring capabilities while preserving general-purpose performance on standard benchmarks. The result is a compact, ownable base that is straightforward to fine-tune, broadly applicable and minimizing the usual capacity–capability trade-offs.

Building on this base, Jan-Code, a code-tuned variant, will be released soon.

Model Overview#

Note: Jan-v3-4B-base-instruct inherits its core architecture from Qwen/Qwen3-4B-Instruct-2507.

  • Number of Parameters: 4.0B
  • Number of Parameters (Non-Embedding): 3.6B
  • Number of Layers: 36
  • Number of Attention Heads (GQA): 32 for Q and 8 for KV
  • Context Length: 262,144 natively.

Intended Use

  • A better small base for downstream work: improved instruction following out of the box, strong starting point for fine-tuning, and effective lightweight coding assistance.

Performance#

Image: image

Quick Start#

Integration with Jan Apps#

Jan-v3 demo is hosted on Jan Browser at chat.jan.ai. It is also optimized for direct integration with Jan Desktop, select the model in the app to start using it.

Local Deployment#

Using vLLM:

vllm serve janhq/Jan-v3-4B-base-instruct \
    --host 0.0.0.0 \
    --port 1234 \
    --enable-auto-tool-choice \
    --tool-call-parser hermes

Using llama.cpp:

llama-server --model Jan-v3-4B-base-instruct-Q8_0.gguf \
    --host 0.0.0.0 \
    --port 1234 \
    --jinja \
    --no-context-shift

For optimal performance in agentic and general tasks, we recommend the following inference parameters:

temperature: 0.7
top_p: 0.8
top_k: 20

🤝 Community & Support#

📄 Citation#

Updated Soon

Files

14 files · 8.84 GBWeights 8.82 GB (safetensors). Sizes and sha256 from the signed manifest.
Hide all 14 files with their sha256
FileSizesha256
.gitattributes1.57 KB34448b82c17d60fec9b65b1f093c115ddbaadc04beb1b0140b6bfed2e012a930
README.md2.85 KB8f3718ddb9bf501c6db2ad21c2449720195d0a49ff0a517451f5cc0d89f46760
added_tokens.json707 Bc0284b582e14987fbd3d5a2cb2bd139084371ed9acbae488829a1c900833c680
chat_template.jinja2.63 KB64f85b198065d0fba2a81f37e10ed68161ce2c19a754c7100e67e0ca2ee9c326
config.json1.54 KB4ca339ac76a475603caf526c2c1f21df1c419e9dfcfae5b512dcad6a34dbe4a1
generation_config.json213 B4d9818c3d27895c0058828a5f68bc7a4de80c3ae1bcdac90936ce24178063f59
merges.txt1.67 MB8831e4f1a044471340f7c0a83d7bd71306a5b867e95fd870f74d0c5308a904d5
model-00001-of-00002.safetensors4.97 GBa28242910de2cda3aa8720c5c7172acf3ee6201d23a41e358848e18173a2b8b7
model-00002-of-00002.safetensors3.85 GBaa3d6f775669cdb0bb5bda638e65d8402bf836c0e34ac91ed602f23b08fe7415
model.safetensors.index.json32.9 KB8744a03005339374bf9de812cc9856bb6af6f5060944fca4b4d25c01ecd2e758
special_tokens_map.json613 B76862e765266b85aa9459767e33cbaf13970f327a0e88d1c65846c2ddd3a1ecd
tokenizer.json11.4 MBaeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
tokenizer_config.json5.41 KB4b5f2f80f84faefe8420e1616671adb1dd3d7e632038d34b1f0e3a1363a51059
vocab.json2.78 MBca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910
VersionsEach time the pinned upstream commit or file list changed in this repo. Newest first.
Upstream commitDateContents
f331713eca pinned now2026-11-15 03:42 UTC14 files · 8.84 GB

Use it

$ fw get janhq/Jan-v3-4B-base-instruct
# manifest sha256 0af9580d8c1a… · 8.84 GB · 14 files
$ fw verify ./Jan-v3-4B-base-instruct janhq/Jan-v3-4B-base-instruct
fw.pyChecks the signature and every sha256, tries each source in order, resumes.
$ python3 fw.py get https://freeweights.allternit.com/manifests/janhq/Jan-v3-4B-base-instruct.json --dir ./Jan-v3-4B-base-instruct $ python3 fw.py verify ./Jan-v3-4B-base-instruct ./janhq__Jan-v3-4B-base-instruct.json
hfThe Hugging Face CLI with one setting. Same repo id.
$ HF_ENDPOINT=https://freeweights.allternit.com hf download janhq/Jan-v3-4B-base-instruct --revision f331713ecaef85c9da95fe5b63131ddb0b8c6744 --local-dir ./Jan-v3-4B-base-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("janhq/Jan-v3-4B-base-instruct", revision="f331713ecaef85c9da95fe5b63131ddb0b8c6744") model = AutoModelForCausalLM.from_pretrained("janhq/Jan-v3-4B-base-instruct", revision="f331713ecaef85c9da95fe5b63131ddb0b8c6744", device_map="auto")
vLLMServe an OpenAI-compatible API.
$ HF_ENDPOINT=https://freeweights.allternit.com vllm serve janhq/Jan-v3-4B-base-instruct --revision f331713ecaef85c9da95fe5b63131ddb0b8c6744
Run at home

License

Name
Apache License 2.0
Id
Apache-2.0
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
8f3718ddb9bf501c6db2ad21c2449720195d0a49ff0a517451f5cc0d89f46760
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: janhq <huggingface.co/janhq>Newsgroups: alt.freeweights.textSubject: Jan-v3-4B-base-instructDate: 2026-11-15Commit: f331713Signed: 2DC28CC7B26A283B ✓
Upstream
janhq/Jan-v3-4B-base-instruct
Pinned commit
f331713ecaef85c9da95fe5b63131ddb0b8c6744
Manifest
Jan-v3-4B-base-instruct.json · .minisig
Manifest sha256
0af9580d8c1aed286c8109e87d24eb2a0129379a492a97deda7376b2b738c0eb
Signature
minisign, Ed25519, key 2DC28CC7B26A283B
Created
2026-11-15 03:42 UTC
Generator
freeweights-manifest/0.1.0
Takedowns
None.

How to verify

$ curl -O https://freeweights.allternit.com/manifests/janhq/Jan-v3-4B-base-instruct.json $ curl -O https://freeweights.allternit.com/manifests/janhq/Jan-v3-4B-base-instruct.json.minisig $ minisign -Vm Jan-v3-4B-base-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

Parent

Qwen/Qwen3-4B-Instruct-2507 · this model is a finetune of it

1 model in the catalog name this one as their parent.

Quantizations
1
Quantizations 1
Get it
Download
$ HF_ENDPOINT=https://freeweights.allternit.com hf download janhq/Jan-v3-4B-base-instruct
Open in Freeweights Studio

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

Signed manifest
minisign, key 2DC28CC7B26A283B. Jan-v3-4B-base-instruct.json. Provenance. Takedown history: none.
Source
huggingface.co/janhq/Jan-v3-4B-base-instruct

Variants

1 quantized or adapter build of Jan-v3-4B-base-instruct. Each has its own license check and page.

SubjectSizeParamsLicenseDate
Jan-v3-4B-base-instruct-4bitHosted (signed) From: mlx-community in alt.freeweights.text2.27 GB4BPermissive2026-11-15

Related models

All of alt.freeweights.text →
From: laion <huggingface.co/laion>Subject: a2-rl-e2egit_large-50-32B
From: laion <huggingface.co/laion>Subject: a3-rl-DCAgent_exp_rpt_e2egit-large-15-8B
From: laion <huggingface.co/laion>Subject: a3-rl-DCAgent_exp_rpt_e2egit-v2-10-8B
From: laion <huggingface.co/laion>Subject: a3-rl-DCAgent_exp_rpt_pymethods2test-large-80-8B
From: laion <huggingface.co/laion>Subject: a3-rl-DCAgent_exp_rpt_unitsyn-python-large-10-8B
From: laion <huggingface.co/laion>Subject: a3-rl-DCAgent_mix_h2_language_proportional-65-8B