CatalogWantedMirrorsLedgerOperating StatementDocsStudio statusCard system
alt.freeweights.text / Trinity-Nano-Base
From: arcee-ai <huggingface.co/arcee-ai>Newsgroups: alt.freeweights.textSubject:

Trinity-Nano-Base

Date: 2026-11-15Commit: 923a95dSigned: 2DC28CC7B26A283B ✓

Text generation model with 6.1B parameters. Safetensors weights, 12.3 GB in all. Published by arcee-ai under OpenMDW-1.1.

6.1B params12.3GBTrinity-Nano-Basesafetensorscustom terms
FactsParams6.1BSize12.3 GBFormatssafetensorsTasktext generationLicenseOpenMDW-1.1From sourcecustom termsread terms
# custom terms: read the license before you download
$ fw get arcee-ai/Trinity-Nano-Base
# manifest sha256 8bcdd2b70629… · 12.3 GB · 20 files
$ fw verify ./Trinity-Nano-Base arcee-ai/Trinity-Nano-Base
From sourceHosted (signed)arcee-ai/Trinity-Nano-BaseLast check: 2026-11-15 03:32 UTC
Run at homeGet help

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

Fine-tune of Trinity-Nano-Base-Pre-Anneal.

Works with

Run24Fine-tune29Convert15Evaluate13Scan13Apps58
OverviewModel cardFilesUse itLicenseProvenanceModel tree

Overview

Size
12.3 GB
Weights
12.2 GB (safetensors)
Parameters
6.1B
Formats
safetensors
Task
text generation
License
OpenMDW-1.1 (custom)
Files
20, each with a sha256
Publisher
arcee-ai
Added
2026-11-15
Section
alt.freeweights.text
Lineage

Original release from the publisher.

Base model: arcee-ai/Trinity-Nano-Base-Pre-Anneal

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 03:32 UTCpassed

Model card

From the publisher at commit 923a95d. Shown as plain text with links; images open at the publisher.

Image: Arcee Trinity Mini

Trinity Nano Base#

Trinity Nano is an Arcee AI 6B MoE model with 1B active parameters. It is the small-sized model in our new Trinity family, a series of open-weight models for enterprise and tinkerers alike.

This base model pre fine tuning, and so is not suitable for chatting, and should be trained for your specific domain before use.

Trinity Nano is trained on 10T tokens gathered and curated through a key partnership with Datology, building upon the excellent dataset we used on AFM-4.5B with additional math and code.

Training was performed on a cluster of 512 H200 GPUs powered by Prime Intellect using HSDP parallelism.

More details, including key architecture decisions, can be found on our blog here

Model Details#

  • Model Architecture: AfmoeForCausalLM
  • Parameters: 6B, 1B active
  • Experts: 128 total, 8 active, 1 shared
  • Context length: 128k
  • Training Tokens: 10T
  • License: OpenMDW-1.1

Benchmarks#

🔢 Math & Reasoning#
BenchmarkScore
GSM8K58.4%
Minerva Math 50036.0%
DROP (0-shot)4.5%
DROP (5-shot)63.6%
💻 Code Generation#
BenchmarkPass@1Pass@10
HumanEval (3-shot, bpb)36.3% (bpb)-
HumanEval+ (temp 0.8)31.7%-
MBPP+44.7%-
🧠 Knowledge & Reasoning#
Benchmark5-shot0-shot
ARC-Challenge84.0%78.2%
ARC-Easy94.8%91.2%
CommonsenseQA74.9%62.7%
OpenBookQA82.2%75.2%
WinoGrande72.8%68.0%
MMLU67.7%64.2%
MMLU Pro35.8%27.7%
AGI Eval (English)51.8%-
BBH (CoT)50.4%7.6%
📘 Understanding & QA#
BenchmarkScore
BoolQ (5-shot)84.3%
HellaSwag (5-shot)77.4%
PIQA (5-shot)82.2%
SciQ (5-shot)93.2%
Social IQA (5-shot)73.0%

Image: Powered by Datology

License#

Trinity-Mini-Base is released under the OpenMDW-1.1 license.

Files

20 files · 12.3 GBWeights 12.2 GB (safetensors). Sizes and sha256 from the signed manifest.
Hide all 20 files with their sha256
FileSizesha256
.gitattributes1.57 KB34448b82c17d60fec9b65b1f093c115ddbaadc04beb1b0140b6bfed2e012a930
LICENSE2.62 KB2ab44b68365473c112f5092211a38f231cb23e50de68b75a13369adbd76a74df
README.md2.95 KBfa343c1767181d32cafa1ab29ec472f30e87c6c44418b25ef65e61e8cc3b2987
__init__.py141 Bcd5a281b63d67a096b2a47a9845f5dd56d6fda4f5b500ca863cd1c26c5f7851b
chat_template.jinja2.71 KBc295e73aea820982584a1f874fa71c61b1f3e6856adc6ef1d7efe339b936f2ad
config.json2.56 KB15f465cef56c5b1fe472784b4c72197ab949056bc73e68a8e451e31470f4b12a
configuration_afmoe.py4.76 KBb16e9f24f9b62db5cc5689ea0854d7cb8e433ed861716adce48eec0073e433ef
generation_config.json69 B49b7b376c9a77281e37b1544e00f6b83ffe1d46bd47ff1c5ee04a7d642ecbe2b
model-00001-of-00007.safetensors2.00 GBab26515f0ffb65114b586e231e6287d72ea4e13679a2c1b7f415321d8c6bea5d
model-00002-of-00007.safetensors2.00 GB771bef2bbb2b18d96cf64155b2a6281f2df28a7d227f2cafdca1eba4a4225d6f
model-00003-of-00007.safetensors2.00 GB93b55c5833c76be5df7621fd12f976a2604cc84e808946a4fc13f688e93eb95e
model-00004-of-00007.safetensors2.00 GB9ec03ea4436bbc13c6edfdd5ecdc517e40359080e20bdf5aad00a26ee40c8578
model-00005-of-00007.safetensors2.00 GBcad464c109dc7c023afbe525f0a2444a3eb7330a01409bfcf7b0bcbef3a445c4
model-00006-of-00007.safetensors1.83 GB695f442f1981104944f0a948e91eea49e08acdd2cf5c2eb41baa4477acd9f397
model-00007-of-00007.safetensors410 MB0102d61f7c6208185a38bcf8d87430ddc4748dbc164be35c394dab4b25074bfa
model.safetensors.index.json1.95 MBa3e0a71236c7cced53fc7b8b11f3a6997bafaf5486adab148b88285f2da73815
modeling_afmoe.py26.6 KB67950e54244545ad8b6016dbfafdf0addb05c398f6afc0e77b621a5693be189c
special_tokens_map.json441 Bbea51ad2e39f89c84a4843f5c29f1680b9b7390ad349395bc0d3c63e17ada5c2
tokenizer.json14.6 MB606b0740c5d8940dbee3268e05beaa3e34f4e43bc7df153681122f8fcc78e134
tokenizer_config.json8.84 KB11f43fc2b40a4a24147e9a2863ebf7f2dbcdb05a28db7edb274c4f90a239819e
VersionsEach time the pinned upstream commit or file list changed in this repo. Newest first.
Upstream commitDateContents
923a95dbba pinned now2026-11-15 03:32 UTC20 files · 12.3 GB

Use it

# custom terms: read the license before you download
$ fw get arcee-ai/Trinity-Nano-Base
# manifest sha256 8bcdd2b70629… · 12.3 GB · 20 files
$ fw verify ./Trinity-Nano-Base arcee-ai/Trinity-Nano-Base
Terms that come with these files. OpenMDW 1.1: keep the agreement and the copyright notices with the files when you pass them on.
Custom license (custom). Read the license text first. Open this panel to show the commands once you accept its terms.
fw.pyChecks the signature and every sha256, tries each source in order, resumes.
$ python3 fw.py get https://freeweights.allternit.com/manifests/arcee-ai/Trinity-Nano-Base.json --dir ./Trinity-Nano-Base $ python3 fw.py verify ./Trinity-Nano-Base ./arcee-ai__Trinity-Nano-Base.json
hfThe Hugging Face CLI with one setting. Same repo id.
$ HF_ENDPOINT=https://freeweights.allternit.com hf download arcee-ai/Trinity-Nano-Base --revision 923a95dbba013c78c4fd6dbc6bce31eea553da7c --local-dir ./Trinity-Nano-Base
transformersLoad the weights in Python.
import os os.environ["HF_ENDPOINT"] = "https://freeweights.allternit.com" from transformers import AutoModelForCausalLM, AutoTokenizer tok = AutoTokenizer.from_pretrained("arcee-ai/Trinity-Nano-Base", revision="923a95dbba013c78c4fd6dbc6bce31eea553da7c") model = AutoModelForCausalLM.from_pretrained("arcee-ai/Trinity-Nano-Base", revision="923a95dbba013c78c4fd6dbc6bce31eea553da7c", device_map="auto")
vLLMServe an OpenAI-compatible API.
$ HF_ENDPOINT=https://freeweights.allternit.com vllm serve arcee-ai/Trinity-Nano-Base --revision 923a95dbba013c78c4fd6dbc6bce31eea553da7c
Run at home

License

Name
OpenMDW-1.1
Id
OpenMDW-1.1
Class
Custom terms
Redistribution
click-through
Listing
Hosted (signed)
Terms
OpenMDW 1.1: keep the agreement and the copyright notices with the files when you pass them on.
Decided
License as declared by the publisher. No person read it. 2026-11-15
Text in
LICENSE
Text sha256
2ab44b68365473c112f5092211a38f231cb23e50de68b75a13369adbd76a74df
Terms gate
Yes. The commands stay folded until you open them after reading 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: arcee-ai <huggingface.co/arcee-ai>Newsgroups: alt.freeweights.textSubject: Trinity-Nano-BaseDate: 2026-11-15Commit: 923a95dSigned: 2DC28CC7B26A283B ✓
Upstream
arcee-ai/Trinity-Nano-Base
Pinned commit
923a95dbba013c78c4fd6dbc6bce31eea553da7c
Manifest
Trinity-Nano-Base.json · .minisig
Manifest sha256
8bcdd2b70629b9cad72368939e92f4492a0aa5204f868112454119fc5358abea
Signature
minisign, Ed25519, key 2DC28CC7B26A283B
Created
2026-11-15 03:32 UTC
Generator
freeweights-manifest/0.1.0
Takedowns
None.

How to verify

$ curl -O https://freeweights.allternit.com/manifests/arcee-ai/Trinity-Nano-Base.json $ curl -O https://freeweights.allternit.com/manifests/arcee-ai/Trinity-Nano-Base.json.minisig $ minisign -Vm Trinity-Nano-Base.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

arcee-ai/Trinity-Nano-Base-Pre-Anneal · this model is a finetune of it

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

Fine-tunes
1
Fine-tunes 1
Get it
Read license, then download
$ HF_ENDPOINT=https://freeweights.allternit.com hf download arcee-ai/Trinity-Nano-Base
Open in Freeweights Studio

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

Signed manifest
minisign, key 2DC28CC7B26A283B. Trinity-Nano-Base.json. Provenance. Takedown history: none.
Source
huggingface.co/arcee-ai/Trinity-Nano-Base

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