CatalogWantedMirrorsLedgerOperating StatementDocsStudio statusCard system
alt.freeweights.text / Laguna-XS-2.1-4bit
From: mlx-community <huggingface.co/mlx-community>Newsgroups: alt.freeweights.textSubject:

Laguna-XS-2.1-4bit

Date: 2026-11-15Commit: c42e0a8Signed: 2DC28CC7B26A283B ✓

Text generation model with 33.4B parameters. Safetensors weights, 18.8 GB in all. Published by mlx-community under OpenMDW-1.1.

33.4B params18.8GBLaguna-XS-2.1-4bitsafetensorscustom terms
FactsParams33.4BSize18.8 GBFormatssafetensorsTasktext generationLicenseOpenMDW-1.1From sourcecustom termsread terms
# custom terms: read the license before you download
$ fw get mlx-community/Laguna-XS-2.1-4bit
# manifest sha256 054c0e144504… · 18.8 GB · 18 files
$ fw verify ./Laguna-XS-2.1-4bit mlx-community/Laguna-XS-2.1-4bit
From sourceHosted (signed)mlx-community/Laguna-XS-2.1-4bitLast check: 2026-11-15 00:21 UTC
Run at homeGet help

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

Quantized of Laguna-XS-2.1.

Works with

Run24Fine-tune29Convert15Evaluate13Scan13Apps58
OverviewModel cardFilesUse itLicenseProvenanceModel tree

Overview

Size
18.8 GB
Weights
18.8 GB (safetensors)
Parameters
33.4B
Formats
safetensors
Task
text generation
License
OpenMDW-1.1 (custom)
Files
18, each with a sha256
Publisher
mlx-community
Added
2026-11-15
Section
alt.freeweights.text
Lineage

Original release from the publisher.

Base model: poolside/Laguna-XS-2.1

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 00:21 UTCpassed

Model card

From the publisher at commit c42e0a8. Shown as plain text with links; images open at the publisher.

Laguna-XS-2.1 (MLX, 4bit)#

Converted from poolside/Laguna-XS-2.1 to MLX format, quantized to 4 bits (group size 64, 4.503 bpw effective).

Notes#

  • Works with mlx-vlm and oMLX (forcing the model's vlm mode). mlx-lm doesn't support the laguna architecture yet — there's an open PR: mlx-lm#1223.
  • Sometimes I got an empty </think> tag at the start of responses, which isn't that common. It won't affect anything tho.

Performance#

Measured with oMLX's benchmark harness on a Macbook Pro M5 Max 128GB 40 GPU (single request, 128 generated tokens):

promptgen tok/sprefill tok/sTTFT mspeak GB
1k126.0279736718.2
4k121.24052101118.8
8k116.63785216518.9
16k109.13122524819.2
32k91.324621331219.8

Variants#

VariantbpwDiskgen tok/s (1k → 32k)
bf161662 GB70.6 → 58.7
8bit8.50033 GB95.4 → 76.7
6bit6.50125 GB102.9 → 80.9
5bit5.50221 GB115.9 → 87.7
4bit (this repo)4.50318 GB126.0 → 91.3
3bit3.50314 GB137.2 → 98.8

Usage#

uvx --from mlx-vlm mlx_vlm.generate --model mlx-community/Laguna-XS-2.1-4bit --prompt "..." --max-tokens 300

License#

OpenMDW-1.1, inherited from the base model.

Files

18 files · 18.8 GBWeights 18.8 GB (safetensors). Sizes and sha256 from the signed manifest.
Hide all 18 files with their sha256
FileSizesha256
.eval_results/swe-bench_pro.yaml168 B6e991a1ccbf1490737f885f874d51a567474a936d2be6374ef85d1d217237723
.eval_results/swe-bench_verified.yaml182 B1aea25f24584692ec3979c5afe5f69d23ad2bfd22c96f7ef14fc3cbf46d3f566
.eval_results/terminal-bench-2.0.yaml183 B642f1e89366800eba95fc3589ade3b8f44fb2a71b7348c4ed290be2399a506ed
.gitattributes1.52 KB11ad7efa24975ee4b0c3c3a38ed18737f0658a5f75a0a96787b576a78a023361
README.md2.00 KB7916152a1f3a59d1f8bad3c2a6b0c7d9215f74f6eed657dfb7c1bedf855f4772
chat_template.jinja6.12 KB5f1bf14725b6a0e662678e7ca66d7b347bbaa436b09db038472045aba37da2f2
config.json15.3 KBc5da07236e90d8cb15611c4e5414b7000cd9ea23f0adc5644a632531b18f45a5
configuration_laguna.py12.8 KB5fde8e57c4082f0a81daf7596672cd636383a71bb668555840f6b7ca489698c7
generation_config.json463 B00380271cff87fdd107f7e9a6ca562ec03527db044faa976883fc3399f8a1463
model-00001-of-00004.safetensors5.24 GB1402abbee7f06e91436d68aa80b54177a676049a96afc1b46e189fec036a909c
model-00002-of-00004.safetensors5.36 GB8616704b15b3c3e7938ec87ae09c1094b667a91f8ef1d59dd93762cd1fa792d3
model-00003-of-00004.safetensors5.27 GBb0c5c8ebd27ecb09b0800c347f80693908b3500fddfbb104f51776dd2d8022d5
model-00004-of-00004.safetensors2.95 GBe0bcc2fe3a78efcfbf21b4828f1235d85fb24278b9681a5ad1b1f3c0b521939f
model.safetensors.index.json173 KB4d3251ee69c8d30968495bb5ccf0f1a33c07947a5660c040888875efef0bdbf2
modeling_laguna.py40.4 KB7b9e55f5bf7d396cb5dfbf82232930017af6a3d70202fd7ebb312a3c229b5ae6
special_tokens_map.json214 B70cd3459fde61761e9440751a590e89a108c09b1803cc7727f5ad1ed1ea6122b
tokenizer.json7.29 MBff04405d2d1e1b6c77a8be25f0fce9371003a558b055c23248d9e8ca1d956d92
tokenizer_config.json429 B1c05d3da3b62a884f66d05c0b68327a69c491bc105ba370cae94d14f3e4f6af9
VersionsEach time the pinned upstream commit or file list changed in this repo. Newest first.
Upstream commitDateContents
c42e0a8f8d pinned now2026-11-15 00:21 UTC18 files · 18.8 GB

Use it

# custom terms: read the license before you download
$ fw get mlx-community/Laguna-XS-2.1-4bit
# manifest sha256 054c0e144504… · 18.8 GB · 18 files
$ fw verify ./Laguna-XS-2.1-4bit mlx-community/Laguna-XS-2.1-4bit
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/mlx-community/Laguna-XS-2.1-4bit.json --dir ./Laguna-XS-2.1-4bit $ python3 fw.py verify ./Laguna-XS-2.1-4bit ./mlx-community__Laguna-XS-2.1-4bit.json
hfThe Hugging Face CLI with one setting. Same repo id.
$ HF_ENDPOINT=https://freeweights.allternit.com hf download mlx-community/Laguna-XS-2.1-4bit --revision c42e0a8f8d504ceacde015a535dcb286d65c8799 --local-dir ./Laguna-XS-2.1-4bit
transformersLoad the weights in Python.
import os os.environ["HF_ENDPOINT"] = "https://freeweights.allternit.com" from transformers import AutoModelForCausalLM, AutoTokenizer tok = AutoTokenizer.from_pretrained("mlx-community/Laguna-XS-2.1-4bit", revision="c42e0a8f8d504ceacde015a535dcb286d65c8799") model = AutoModelForCausalLM.from_pretrained("mlx-community/Laguna-XS-2.1-4bit", revision="c42e0a8f8d504ceacde015a535dcb286d65c8799", device_map="auto")
vLLMServe an OpenAI-compatible API.
$ HF_ENDPOINT=https://freeweights.allternit.com vllm serve mlx-community/Laguna-XS-2.1-4bit --revision c42e0a8f8d504ceacde015a535dcb286d65c8799
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
README.md
Text sha256
7916152a1f3a59d1f8bad3c2a6b0c7d9215f74f6eed657dfb7c1bedf855f4772
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: mlx-community <huggingface.co/mlx-community>Newsgroups: alt.freeweights.textSubject: Laguna-XS-2.1-4bitDate: 2026-11-15Commit: c42e0a8Signed: 2DC28CC7B26A283B ✓
Upstream
mlx-community/Laguna-XS-2.1-4bit
Pinned commit
c42e0a8f8d504ceacde015a535dcb286d65c8799
Manifest
Laguna-XS-2.1-4bit.json · .minisig
Manifest sha256
054c0e14450460b2f96c657844584821cf882397e19eb38458190a4627137677
Signature
minisign, Ed25519, key 2DC28CC7B26A283B
Created
2026-11-15 00:21 UTC
Generator
freeweights-manifest/0.1.0
Takedowns
None.

How to verify

$ curl -O https://freeweights.allternit.com/manifests/mlx-community/Laguna-XS-2.1-4bit.json $ curl -O https://freeweights.allternit.com/manifests/mlx-community/Laguna-XS-2.1-4bit.json.minisig $ minisign -Vm Laguna-XS-2.1-4bit.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

poolside/Laguna-XS-2.1 · this model is a quantization of it

No model in the catalog names this one as its parent.

Get it
Read license, then download
$ HF_ENDPOINT=https://freeweights.allternit.com hf download mlx-community/Laguna-XS-2.1-4bit
Open in Freeweights Studio

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

Signed manifest
minisign, key 2DC28CC7B26A283B. Laguna-XS-2.1-4bit.json. Provenance. Takedown history: none.
Source
huggingface.co/mlx-community/Laguna-XS-2.1-4bit

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