CatalogWantedMirrorsLedgerOperating StatementDocsStudio statusCard system
alt.freeweights.lfm / LiquidAI_LFM2-1.2B-RAG-GGUF
From: bartowski <huggingface.co/bartowski>Newsgroups: alt.freeweights.lfmSubject:

LiquidAI_LFM2-1.2B-RAG-GGUF

Date: 2026-11-15Commit: d503185Signed: 2DC28CC7B26A283B ✓

Text generation model. Gguf weights, 18.6 GB in all. Published by bartowski under LFM-1.0 (custom).

lfm18.6GBLiquidAI_LFM2-1.2B-RAG-GGUFggufcustom terms
FactsParamsnot statedSize18.6 GBFormatsggufTasktext generationLicenseLFM-1.0 (custom)From sourcecustom termsread terms
# custom terms: read the license before you download
$ fw get bartowski/LiquidAI_LFM2-1.2B-RAG-GGUF
# manifest sha256 219ae7752a73… · 18.6 GB · 27 files
$ fw verify ./LiquidAI_LFM2-1.2B-RAG-GGUF bartowski/LiquidAI_LFM2-1.2B-RAG-GGUF
From sourceHosted (signed)bartowski/LiquidAI_LFM2-1.2B-RAG-GGUFLast check: 2026-11-15 00:16 UTC
Run at homeGet help

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

Quantized of LFM2-1.2B-RAG.

Works with

Run14Convert1Evaluate13Scan12Apps58

Convert with

ggml
OverviewModel cardFilesUse itLicenseProvenanceModel tree

Overview

Size
18.6 GB
Weights
1.16 MB (smallest of 25 GGUF variants)
Parameters
not stated
Formats
gguf
Task
text generation
License
LFM-1.0 (custom) (custom)
Files
27, each with a sha256
Publisher
bartowski
Added
2026-11-15
Section
alt.freeweights.lfm
Lineage

Original release from the publisher.

Base model: LiquidAI/LFM2-1.2B-RAG

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 00:16 UTCpassed

Model card

From the publisher at commit d503185. Shown as plain text with links; images open at the publisher.

Llamacpp imatrix Quantizations of LFM2-1.2B-RAG by LiquidAI#

Using llama.cpp release b6683 for quantization.

Original model: https://huggingface.co/LiquidAI/LFM2-1.2B-RAG

All quants made using imatrix option with dataset from here combined with a subset of combined_all_small.parquet from Ed Addario here

Run them in LM Studio

Run them directly with llama.cpp, or any other llama.cpp based project

Prompt format#

No prompt format found, check original model page

Download a file (not the whole branch) from below:#

FilenameQuant typeFile SizeSplitDescription
LFM2-1.2B-RAG-bf16.ggufbf162.34GBfalseFull BF16 weights.
LFM2-1.2B-RAG-Q8_0.ggufQ8_01.25GBfalseExtremely high quality, generally unneeded but max available quant.
LFM2-1.2B-RAG-Q6_K_L.ggufQ6_K_L1.00GBfalseUses Q8_0 for embed and output weights. Very high quality, near perfect, recommended.
LFM2-1.2B-RAG-Q6_K.ggufQ6_K0.96GBfalseVery high quality, near perfect, recommended.
LFM2-1.2B-RAG-Q5_K_L.ggufQ5_K_L0.88GBfalseUses Q8_0 for embed and output weights. High quality, recommended.
LFM2-1.2B-RAG-Q5_K_M.ggufQ5_K_M0.84GBfalseHigh quality, recommended.
LFM2-1.2B-RAG-Q5_K_S.ggufQ5_K_S0.83GBfalseHigh quality, recommended.
LFM2-1.2B-RAG-Q4_K_L.ggufQ4_K_L0.76GBfalseUses Q8_0 for embed and output weights. Good quality, recommended.
LFM2-1.2B-RAG-Q4_1.ggufQ4_10.76GBfalseLegacy format, similar performance to Q4_K_S but with improved tokens/watt on Apple silicon.
LFM2-1.2B-RAG-Q4_K_M.ggufQ4_K_M0.73GBfalseGood quality, default size for most use cases, recommended.
LFM2-1.2B-RAG-Q4_K_S.ggufQ4_K_S0.70GBfalseSlightly lower quality with more space savings, recommended.
LFM2-1.2B-RAG-Q4_0.ggufQ4_00.70GBfalseLegacy format, offers online repacking for ARM and AVX CPU inference.
LFM2-1.2B-RAG-IQ4_NL.ggufIQ4_NL0.70GBfalseSimilar to IQ4_XS, but slightly larger. Offers online repacking for ARM CPU inference.
LFM2-1.2B-RAG-Q3_K_XL.ggufQ3_K_XL0.67GBfalseUses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability.
LFM2-1.2B-RAG-IQ4_XS.ggufIQ4_XS0.66GBfalseDecent quality, smaller than Q4_K_S with similar performance, recommended.
LFM2-1.2B-RAG-Q3_K_L.ggufQ3_K_L0.64GBfalseLower quality but usable, good for low RAM availability.
LFM2-1.2B-RAG-Q3_K_M.ggufQ3_K_M0.60GBfalseLow quality.
LFM2-1.2B-RAG-IQ3_M.ggufIQ3_M0.57GBfalseMedium-low quality, new method with decent performance comparable to Q3_K_M.
LFM2-1.2B-RAG-Q3_K_S.ggufQ3_K_S0.56GBfalseLow quality, not recommended.
LFM2-1.2B-RAG-IQ3_XS.ggufIQ3_XS0.54GBfalseLower quality, new method with decent performance, slightly better than Q3_K_S.
LFM2-1.2B-RAG-Q2_K_L.ggufQ2_K_L0.52GBfalseUses Q8_0 for embed and output weights. Very low quality but surprisingly usable.
LFM2-1.2B-RAG-IQ3_XXS.ggufIQ3_XXS0.49GBfalseLower quality, new method with decent performance, comparable to Q3 quants.
LFM2-1.2B-RAG-Q2_K.ggufQ2_K0.48GBfalseVery low quality but surprisingly usable.
LFM2-1.2B-RAG-IQ2_M.ggufIQ2_M0.43GBfalseRelatively low quality, uses SOTA techniques to be surprisingly usable.

Embed/output weights#

Some of these quants (Q3_K_XL, Q4_K_L etc) are the standard quantization method with the embeddings and output weights quantized to Q8_0 instead of what they would normally default to.

Downloading using huggingface-cli#

Click to view download instructions

First, make sure you have hugginface-cli installed:

pip install -U "huggingface_hub[cli]"

Then, you can target the specific file you want:

huggingface-cli download bartowski/LiquidAI_LFM2-1.2B-RAG-GGUF --include "LiquidAI_LFM2-1.2B-RAG-Q4_K_M.gguf" --local-dir ./

If the model is bigger than 50GB, it will have been split into multiple files. In order to download them all to a local folder, run:

huggingface-cli download bartowski/LiquidAI_LFM2-1.2B-RAG-GGUF --include "LiquidAI_LFM2-1.2B-RAG-Q8_0/*" --local-dir ./

You can either specify a new local-dir (LiquidAI_LFM2-1.2B-RAG-Q8_0) or download them all in place (./)

ARM/AVX information#

Previously, you would download Q4_0_4_4/4_8/8_8, and these would have their weights interleaved in memory in order to improve performance on ARM and AVX machines by loading up more data in one pass.

Now, however, there is something called "online repacking" for weights. details in this PR. If you use Q4_0 and your hardware would benefit from repacking weights, it will do it automatically on the fly.

As of llama.cpp build b4282 you will not be able to run the Q4_0_X_X files and will instead need to use Q4_0.

Additionally, if you want to get slightly better quality for , you can use IQ4_NL thanks to this PR which will also repack the weights for ARM, though only the 4_4 for now. The loading time may be slower but it will result in an overall speed incrase.

Click to view Q4_0_X_X information (deprecated

I'm keeping this section to show the potential theoretical uplift in performance from using the Q4_0 with online repacking.

Click to view benchmarks on an AVX2 system (EPYC7702)

modelsizeparamsbackendthreadstestt/s% (vs Q4_0)
qwen2 3B Q4_01.70 GiB3.09 BCPU64pp512204.03 ± 1.03100%
qwen2 3B Q4_01.70 GiB3.09 BCPU64pp1024282.92 ± 0.19100%
qwen2 3B Q4_01.70 GiB3.09 BCPU64pp2048259.49 ± 0.44100%
qwen2 3B Q4_01.70 GiB3.09 BCPU64tg12839.12 ± 0.27100%
qwen2 3B Q4_01.70 GiB3.09 BCPU64tg25639.31 ± 0.69100%
qwen2 3B Q4_01.70 GiB3.09 BCPU64tg51240.52 ± 0.03100%
qwen2 3B Q4_K_M1.79 GiB3.09 BCPU64pp512301.02 ± 1.74147%
qwen2 3B Q4_K_M1.79 GiB3.09 BCPU64pp1024287.23 ± 0.20101%
qwen2 3B Q4_K_M1.79 GiB3.09 BCPU64pp2048262.77 ± 1.81101%
qwen2 3B Q4_K_M1.79 GiB3.09 BCPU64tg12818.80 ± 0.9948%
qwen2 3B Q4_K_M1.79 GiB3.09 BCPU64tg25624.46 ± 3.0483%
qwen2 3B Q4_K_M1.79 GiB3.09 BCPU64tg51236.32 ± 3.5990%
qwen2 3B Q4_0_8_81.69 GiB3.09 BCPU64pp512271.71 ± 3.53133%
qwen2 3B Q4_0_8_81.69 GiB3.09 BCPU64pp1024279.86 ± 45.63100%
qwen2 3B Q4_0_8_81.69 GiB3.09 BCPU64pp2048320.77 ± 5.00124%
qwen2 3B Q4_0_8_81.69 GiB3.09 BCPU64tg12843.51 ± 0.05111%
qwen2 3B Q4_0_8_81.69 GiB3.09 BCPU64tg25643.35 ± 0.09110%
qwen2 3B Q4_0_8_81.69 GiB3.09 BCPU64tg51242.60 ± 0.31105%

Q4_0_8_8 offers a nice bump to prompt processing and a small bump to text generation

Which file should I choose?#

Click here for details

A great write up with charts showing various performances is provided by Artefact2 here

The first thing to figure out is how big a model you can run. To do this, you'll need to figure out how much RAM and/or VRAM you have.

If you want your model running as FAST as possible, you'll want to fit the whole thing on your GPU's VRAM. Aim for a quant with a file size 1-2GB smaller than your GPU's total VRAM.

If you want the absolute maximum quality, add both your system RAM and your GPU's VRAM together, then similarly grab a quant with a file size 1-2GB Smaller than that total.

Next, you'll need to decide if you want to use an 'I-quant' or a 'K-quant'.

If you don't want to think too much, grab one of the K-quants. These are in format 'QX_K_X', like Q5_K_M.

If you want to get more into the weeds, you can check out this extremely useful feature chart:

llama.cpp feature matrix

But basically, if you're aiming for below Q4, and you're running cuBLAS (Nvidia) or rocBLAS (AMD), you should look towards the I-quants. These are in format IQX_X, like IQ3_M. These are newer and offer better performance for their size.

These I-quants can also be used on CPU, but will be slower than their K-quant equivalent, so speed vs performance is a tradeoff you'll have to decide.

Credits#

Thank you kalomaze and Dampf for assistance in creating the imatrix calibration dataset.

Thank you ZeroWw for the inspiration to experiment with embed/output.

Thank you to LM Studio for sponsoring my work.

Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski

Files

27 files · 18.6 GBWeights 1.16 MB (smallest of 25 GGUF variants). Sizes and sha256 from the signed manifest.
Show all 27 files with their sha256
FileSizesha256
.gitattributes3.28 KBff57ea1bec7af148399401431f5e3ecd5667b1d30de749f6d821db504cd37477
LiquidAI_LFM2-1.2B-RAG-IQ2_M.gguf434 MB8f024f3629ca799c6edd26708eb14b7bd36c8bef5377db9f4f1e1acac3471d3a
LiquidAI_LFM2-1.2B-RAG-IQ3_M.gguf567 MB7727621c71ae02c60d2739fd79d7c97a8bbced1687fa474a171d08f3d901c959
LiquidAI_LFM2-1.2B-RAG-IQ3_XS.gguf538 MB134cc33df1be0eb31f0a503e6e5f45ae77512a902e0b80872318fd8a3d63cb20
LiquidAI_LFM2-1.2B-RAG-IQ3_XXS.gguf491 MB5ac8cf32003a6986e7afec49c3539262529b586e28869d7c6d89ced5bc8d4ab8
LiquidAI_LFM2-1.2B-RAG-IQ4_NL.gguf696 MB620fc2846f75efef0586118c1b6cb66035af2b1eee2ed059cc578aaceadbe1e4
LiquidAI_LFM2-1.2B-RAG-IQ4_XS.gguf663 MBfb67190b93dfb85c5f0f72b626b7c3300b349f411c5b0f7d441c8937a1e4094b
LiquidAI_LFM2-1.2B-RAG-Q2_K.gguf483 MBf8d59469e47d6aab83f7d323b5f39772ebecfeb3505b452d26bbf6d46cba3906
LiquidAI_LFM2-1.2B-RAG-Q2_K_L.gguf516 MB0c44a59acee8798c34b34499435bdf257e539fb372a8dd4ee52e8100f381e923
LiquidAI_LFM2-1.2B-RAG-Q3_K_L.gguf635 MB72a30b232e5e2b6c21b3e91b282a820393793810bed519c2615da60959dfe1f3
LiquidAI_LFM2-1.2B-RAG-Q3_K_M.gguf600 MBbe40630db52b2dd55d460c0292c12af75fe36ccd683c576d1aec6b2c00c682ba
LiquidAI_LFM2-1.2B-RAG-Q3_K_S.gguf558 MB65de5236f5a60dae773c6daa81764f94f96479e5a1d187175a0acbb0d0654c0e
LiquidAI_LFM2-1.2B-RAG-Q3_K_XL.gguf668 MBb09c7e12e2d6a28adee5f0c93ed0624f8e251423674605db43e3b50d147e3387
LiquidAI_LFM2-1.2B-RAG-Q4_0.gguf698 MBf2653f84b003e1bdecc42d6c15fd7ef50f13fbc242ca29a6968e48f82d42feea
LiquidAI_LFM2-1.2B-RAG-Q4_1.gguf761 MBcdbbcdf8b6054c288acca03fc29045ad0e7f34b5c266dd9c9f7803be7de2998f
LiquidAI_LFM2-1.2B-RAG-Q4_K_L.gguf763 MB9b13a9d4c351010f21808586363fa9d4c26933b8fcb831ea4416eb54d87f2ecf
LiquidAI_LFM2-1.2B-RAG-Q4_K_M.gguf731 MB11c93b5ae81612ab532fcfb395fddd2fb478b5d6215e1b46eeee3576a31eaa2d
LiquidAI_LFM2-1.2B-RAG-Q4_K_S.gguf700 MB75eaa025ad2a28504f3ff371ec8d347f19d47c8622fd74047110cef5ccd2bed6
LiquidAI_LFM2-1.2B-RAG-Q5_K_L.gguf876 MB369a73729a6c014b7b6a45b4fe93998f9d8b4dd50279abfee483bacb1d5e9de5
LiquidAI_LFM2-1.2B-RAG-Q5_K_M.gguf843 MB3b4a1ad507f1c1a2bfb569448f56e59a1b1c3f944f532c331411b27fcdf1067c
LiquidAI_LFM2-1.2B-RAG-Q5_K_S.gguf825 MB6b9590bcde01dceb0eb8d25cc627da4a88facce361c1cbc7a037f2c7bcbf7d74
LiquidAI_LFM2-1.2B-RAG-Q6_K.gguf963 MB2d792b9134f51de0ea222bf8164ceef2fa922f9c8b5ea6d693c0b300cb7fc2bf
LiquidAI_LFM2-1.2B-RAG-Q6_K_L.gguf995 MB081c523642b00b04d09b6fcd7ac2fc497ede0ebff59d99ef37fffae90246ba3a
LiquidAI_LFM2-1.2B-RAG-Q8_0.gguf1.25 GB8b5c536842368b17a517274087e8e75a93d2b34aa921669abb4cda3ca2912cfe
LiquidAI_LFM2-1.2B-RAG-bf16.gguf2.34 GBc55775b5b61fc70837724f644fd41d8aac6b9d89ffacf32e7390eb541120db51
LiquidAI_LFM2-1.2B-RAG-imatrix.gguf1.16 MB3d932eeab8f9d062bf43b1424e00ca5a413256fe7205803e109ad75a3de72f94
README.md13.6 KB1dc42a3a5b2c75e304c55a0ab337588e4a22c81478b5410ea52051c77526d8f6
VersionsEach time the pinned upstream commit or file list changed in this repo. Newest first.
Upstream commitDateContents
d503185689 pinned now2026-11-15 00:16 UTC27 files · 18.6 GB

Use it

# custom terms: read the license before you download
$ fw get bartowski/LiquidAI_LFM2-1.2B-RAG-GGUF
# manifest sha256 219ae7752a73… · 18.6 GB · 27 files
$ fw verify ./LiquidAI_LFM2-1.2B-RAG-GGUF bartowski/LiquidAI_LFM2-1.2B-RAG-GGUF
Terms that come with these files. LFM 1.0: commercial use only for organisations under US$10M annual revenue. Keep a copy of the license with the files.
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/bartowski/LiquidAI_LFM2-1.2B-RAG-GGUF.json --dir ./LiquidAI_LFM2-1.2B-RAG-GGUF $ python3 fw.py verify ./LiquidAI_LFM2-1.2B-RAG-GGUF ./bartowski__LiquidAI_LFM2-1.2B-RAG-GGUF.json
hfThe Hugging Face CLI with one setting. Same repo id.
$ HF_ENDPOINT=https://freeweights.allternit.com hf download bartowski/LiquidAI_LFM2-1.2B-RAG-GGUF --revision d5031856893983071e2d131fc7c897e4e25afcb9 --local-dir ./LiquidAI_LFM2-1.2B-RAG-GGUF
llama.cppDownload one GGUF variant (LiquidAI_LFM2-1.2B-RAG-imatrix, 1.16 MB), then serve it.
$ HF_ENDPOINT=https://freeweights.allternit.com hf download bartowski/LiquidAI_LFM2-1.2B-RAG-GGUF --include "LiquidAI_LFM2-1.2B-RAG-imatrix*" --local-dir ./LiquidAI_LFM2-1.2B-RAG-GGUF $ llama-server -m ./LiquidAI_LFM2-1.2B-RAG-GGUF/LiquidAI_LFM2-1.2B-RAG-imatrix.gguf
OllamaMake an Ollama model from the downloaded file.
$ printf 'FROM ./LiquidAI_LFM2-1.2B-RAG-GGUF/LiquidAI_LFM2-1.2B-RAG-imatrix.gguf\n' > Modelfile $ ollama create liquidai-lfm2-1.2b-rag-gguf -f Modelfile
Run at home

License

Name
LFM Open License v1.0
Id
LicenseRef-LFM-1.0
Class
Custom terms
Redistribution
click-through
Listing
Hosted (signed)
Terms
LFM 1.0: commercial use only for organisations under US$10M annual revenue. Keep a copy of the license with the files.
Decided
Inherited from the base model LiquidAI/LFM2-1.2B-RAG: the strictest license along the lineage applies. 2026-11-15
Text in
README.md
Text sha256
1dc42a3a5b2c75e304c55a0ab337588e4a22c81478b5410ea52051c77526d8f6
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: bartowski <huggingface.co/bartowski>Newsgroups: alt.freeweights.lfmSubject: LiquidAI_LFM2-1.2B-RAG-GGUFDate: 2026-11-15Commit: d503185Signed: 2DC28CC7B26A283B ✓
Upstream
bartowski/LiquidAI_LFM2-1.2B-RAG-GGUF
Pinned commit
d5031856893983071e2d131fc7c897e4e25afcb9
Manifest
LiquidAI_LFM2-1.2B-RAG-GGUF.json · .minisig
Manifest sha256
219ae7752a73d80d133906242765e95d6156038125dbb98ddf8476d2290cfb12
Signature
minisign, Ed25519, key 2DC28CC7B26A283B
Created
2026-11-15 00:16 UTC
Generator
freeweights-manifest/0.1.0
Takedowns
None.

How to verify

$ curl -O https://freeweights.allternit.com/manifests/bartowski/LiquidAI_LFM2-1.2B-RAG-GGUF.json $ curl -O https://freeweights.allternit.com/manifests/bartowski/LiquidAI_LFM2-1.2B-RAG-GGUF.json.minisig $ minisign -Vm LiquidAI_LFM2-1.2B-RAG-GGUF.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

LiquidAI/LFM2-1.2B-RAG · 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 bartowski/LiquidAI_LFM2-1.2B-RAG-GGUF
Open in Freeweights Studio

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

Signed manifest
minisign, key 2DC28CC7B26A283B. LiquidAI_LFM2-1.2B-RAG-GGUF.json. Provenance. Takedown history: none.
Source
huggingface.co/bartowski/LiquidAI_LFM2-1.2B-RAG-GGUF

Related models

All of alt.freeweights.lfm →
From: LiquidAI <huggingface.co/LiquidAI>Subject: LFM2-1.2B
From: LiquidAI <huggingface.co/LiquidAI>Subject: LFM2-1.2B-Extract
From: LiquidAI <huggingface.co/LiquidAI>Subject: LFM2-1.2B-RAG
From: LiquidAI <huggingface.co/LiquidAI>Subject: LFM2-1.2B-Tool
From: LiquidAI <huggingface.co/LiquidAI>Subject: LFM2-2.6B
From: LiquidAI <huggingface.co/LiquidAI>Subject: LFM2-2.6B-Exp