alt.freeweights.mistral / MegaBeam-Mistral-7B-512k-GGUF

From: bartowski <huggingface.co/bartowski>Newsgroups: alt.freeweights.mistralSubject:

MegaBeam-Mistral-7B-512k-GGUF

Date: 2026-11-15Commit: f6a596eSigned: 2DC28CC7B26A283B ✓

Text generation model. Gguf weights, 95.2 GB in all. Published by bartowski under Apache-2.0.

mistral95.2GBMegaBeam-Mistral-7B-512k-GGUFggufpermissive

FactsParamsnot statedSize95.2 GBFormatsggufTasktext generationLicenseApache-2.0From sourcepermissive

$ fw get bartowski/MegaBeam-Mistral-7B-512k-GGUF
# manifest sha256 76c86376466d… · 95.2 GB · 23 files
$ fw verify ./MegaBeam-Mistral-7B-512k-GGUF bartowski/MegaBeam-Mistral-7B-512k-GGUF

From sourceHosted (signed)bartowski/MegaBeam-Mistral-7B-512k-GGUFLast check: 2026-11-15 04:55 UTC

Run at home Get help

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

Quantized of MegaBeam-Mistral-7B-512k.

Works with

Run14 Convert1 Evaluate13 Scan12 Apps58

Runs with

Ranked by fit to this model (gguf), then by GitHub stars.

Text Generation WebUI mistral.rs

Ollama

llama.cpp LocalAI

BitNet llamafile gpu-unlock KoboldCpp PocketPal AI llama-swap ik_llama.cpp

Show all 14

RamaLama Maid

Convert with

ggml

Evaluate with

Opik DeepEval Ragas lm-evaluation-harness OpenCompass SWE-bench

simple-evals TruLens HELM

LightEval

Evaluate Weave

Show all 13

Autoevals

Scan with

promptfoo garak Guardrails NeMo Guardrails Giskard DeepTeam AI Fairness 360 Fairlearn Detoxify MCP Scanner AgentDojo ModelScan

Overview Model card Files Use it License Provenance Model tree

Overview

Size: 95.2 GB
Weights: 2.50 GB (smallest of 20 GGUF variants)
Parameters: not stated
Formats: gguf
Task: text generation
License: Apache-2.0 (permissive)
Files: 23, each with a sha256
Publisher: bartowski
Added: 2026-11-15
Section: alt.freeweights.mistral

Lineage

Original release from the publisher.

Base model: asp-prototyping/MegaBeam-Mistral-7B-512k

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

Badge	Host	Last check	Result
From source	huggingface.co	2026-11-15 04:55 UTC	passed

Model card

From the publisher at commit f6a596e. Shown as plain text with links; images open at the publisher.

Llamacpp imatrix Quantizations of MegaBeam-Mistral-7B-512k#

Using llama.cpp release b3600 for quantization.

Original model: https://huggingface.co/aws-prototyping/MegaBeam-Mistral-7B-512k

All quants made using imatrix option with dataset from here

Run them in LM Studio

Prompt format#

<s> [INST] {prompt} [/INST]

Note that this model does not support a System prompt.

What's new:#

Model updated for "improved user experience" and fixing repetition issues

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

Filename	Quant type	File Size	Split	Description
MegaBeam-Mistral-7B-512k-f16.gguf	f16	14.48GB	false	Full F16 weights.
MegaBeam-Mistral-7B-512k-Q8_0.gguf	Q8_0	7.70GB	false	Extremely high quality, generally unneeded but max available quant.
MegaBeam-Mistral-7B-512k-Q6_K_L.gguf	Q6_K_L	6.01GB	false	Uses Q8_0 for embed and output weights. Very high quality, near perfect, recommended.
MegaBeam-Mistral-7B-512k-Q6_K.gguf	Q6_K	5.94GB	false	Very high quality, near perfect, recommended.
MegaBeam-Mistral-7B-512k-Q5_K_L.gguf	Q5_K_L	5.21GB	false	Uses Q8_0 for embed and output weights. High quality, recommended.
MegaBeam-Mistral-7B-512k-Q5_K_M.gguf	Q5_K_M	5.13GB	false	High quality, recommended.
MegaBeam-Mistral-7B-512k-Q5_K_S.gguf	Q5_K_S	5.00GB	false	High quality, recommended.
MegaBeam-Mistral-7B-512k-Q4_K_L.gguf	Q4_K_L	4.47GB	false	Uses Q8_0 for embed and output weights. Good quality, recommended.
MegaBeam-Mistral-7B-512k-Q4_K_M.gguf	Q4_K_M	4.37GB	false	Good quality, default size for must use cases, recommended.
MegaBeam-Mistral-7B-512k-Q4_K_S.gguf	Q4_K_S	4.14GB	false	Slightly lower quality with more space savings, recommended.
MegaBeam-Mistral-7B-512k-Q3_K_XL.gguf	Q3_K_XL	3.94GB	false	Uses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability.
MegaBeam-Mistral-7B-512k-IQ4_XS.gguf	IQ4_XS	3.91GB	false	Decent quality, smaller than Q4_K_S with similar performance, recommended.
MegaBeam-Mistral-7B-512k-Q3_K_L.gguf	Q3_K_L	3.82GB	false	Lower quality but usable, good for low RAM availability.
MegaBeam-Mistral-7B-512k-Q3_K_M.gguf	Q3_K_M	3.52GB	false	Low quality.
MegaBeam-Mistral-7B-512k-IQ3_M.gguf	IQ3_M	3.28GB	false	Medium-low quality, new method with decent performance comparable to Q3_K_M.
MegaBeam-Mistral-7B-512k-Q3_K_S.gguf	Q3_K_S	3.16GB	false	Low quality, not recommended.
MegaBeam-Mistral-7B-512k-IQ3_XS.gguf	IQ3_XS	3.02GB	false	Lower quality, new method with decent performance, slightly better than Q3_K_S.
MegaBeam-Mistral-7B-512k-Q2_K_L.gguf	Q2_K_L	2.85GB	false	Uses Q8_0 for embed and output weights. Very low quality but surprisingly usable.
MegaBeam-Mistral-7B-512k-Q2_K.gguf	Q2_K	2.72GB	false	Very low quality but surprisingly usable.
MegaBeam-Mistral-7B-512k-IQ2_M.gguf	IQ2_M	2.50GB	false	Relatively 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.

Some say that this improves the quality, others don't notice any difference. If you use these models PLEASE COMMENT with your findings. I would like feedback that these are actually used and useful so I don't keep uploading quants no one is using.

Thanks!

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

Downloading using huggingface-cli#

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/MegaBeam-Mistral-7B-512k-GGUF --include "MegaBeam-Mistral-7B-512k-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/MegaBeam-Mistral-7B-512k-GGUF --include "MegaBeam-Mistral-7B-512k-Q8_0/*" --local-dir ./

You can either specify a new local-dir (MegaBeam-Mistral-7B-512k-Q8_0) or download them all in place (./)

Which file should I choose?#

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 and Apple Metal, but will be slower than their K-quant equivalent, so speed vs performance is a tradeoff you'll have to decide.

The I-quants are not compatible with Vulcan, which is also AMD, so if you have an AMD card double check if you're using the rocBLAS build or the Vulcan build. At the time of writing this, LM Studio has a preview with ROCm support, and other inference engines have specific builds for ROCm.

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

Files

23 files · 95.2 GBWeights 2.50 GB (smallest of 20 GGUF variants). Sizes and sha256 from the signed manifest.

Show all 23 files with their sha256

File	Size	sha256
.gitattributes	3.11 KB	c10a1c277d78bedf11b1f46204c208e9de595433454fb0fabba70659652c99da
MegaBeam-Mistral-7B-512k-IQ2_M.gguf	2.50 GB	13a070156ca0d9fb155d1868f9cb8ab05a58102da3ba472ab0a093a60ece6afa
MegaBeam-Mistral-7B-512k-IQ3_M.gguf	3.28 GB	32993764913c9a5be2b5451cd427efb0f093f943d5747a81cbfaec67d2eb30b4
MegaBeam-Mistral-7B-512k-IQ3_XS.gguf	3.02 GB	6da6551ec1c3175b7650bd9fc4e421730383c2566adfc1e4e79b401b98c42e94
MegaBeam-Mistral-7B-512k-IQ4_XS.gguf	3.91 GB	ce571e5d43dc5f7f92dde14f2f468387e9dcc1fc9bfaf686f75e7deae848efc1
MegaBeam-Mistral-7B-512k-Q2_K.gguf	2.72 GB	6f25509de0d9f1a658ce4686e2f6b784c9bdf1358607159d30cec9d3b4d7f52b
MegaBeam-Mistral-7B-512k-Q2_K_L.gguf	2.85 GB	0ac4e21acb43ec846a518212901059bd462794fe4980df9534c702f048b27a3b
MegaBeam-Mistral-7B-512k-Q3_K_L.gguf	3.82 GB	074fcd41a1f5a4cfcf092de21b19c244950d61744bc075c3184d3846464b701e
MegaBeam-Mistral-7B-512k-Q3_K_M.gguf	3.52 GB	450fd6994603c9ead29d17eda50cc25fbe77be84cb0741122279bbda0728fe8d
MegaBeam-Mistral-7B-512k-Q3_K_S.gguf	3.16 GB	2d091a6ad2cbb8708be977bfe4eb31980005361f584b75d1cf89386ed2577d1e
MegaBeam-Mistral-7B-512k-Q3_K_XL.gguf	3.94 GB	0da322eff85a46660daa1b1aeb22623b9854ec7795bc65253d2183e0d6d4ff5d
MegaBeam-Mistral-7B-512k-Q4_K_L.gguf	4.47 GB	829357e8a5a7a721d6c7cdcb4bbf7696ec6e914608bde6e5498905defe19cf82
MegaBeam-Mistral-7B-512k-Q4_K_M.gguf	4.37 GB	7bba5d61ffda53ae6df7c4001d825fa82bcf9cbb769ae3b4ec7f9fd3c291d753
MegaBeam-Mistral-7B-512k-Q4_K_S.gguf	4.14 GB	c4b993d4ede942a35118ed6c677e8579a0e533c610ebd6572f71e68186085ea5
MegaBeam-Mistral-7B-512k-Q5_K_L.gguf	5.21 GB	0d7b75f5ed535a918cd3b93c2f9bea8e0c1ae003234e8cc5a74ec960a41e0711
MegaBeam-Mistral-7B-512k-Q5_K_M.gguf	5.13 GB	8447ebea5eeb5e8b0d5eed4df43af095adfbc8f66b014f4ea9f50b95b5c90663
MegaBeam-Mistral-7B-512k-Q5_K_S.gguf	5.00 GB	84571aa5e7d9e5bf9e1086fd06555efaa0cc180529f26fbe5ba8b0a7dcda0ea5
MegaBeam-Mistral-7B-512k-Q6_K.gguf	5.94 GB	9d139c0fb07db1c239d3367cd49dd17c9886214cb9b27a6e777609ada92ca270
MegaBeam-Mistral-7B-512k-Q6_K_L.gguf	6.01 GB	1cf982115a7c820918ca681a84bd48116ddbd82fc8ab2a6e085da6acbb7abc79
MegaBeam-Mistral-7B-512k-Q8_0.gguf	7.70 GB	c99568eee9732640dc69b8fb564a3ce3aec528744acbf1d0c23c56d47d9ee81a
MegaBeam-Mistral-7B-512k-f16.gguf	14.5 GB	83c2a4ce8ecaec28cc92b2c8cef39f38f5e4abdb97dad436247962308a4f8075
MegaBeam-Mistral-7B-512k.imatrix	4.99 MB	406b73017d5bbd0e6b7ad0f94ff71287e0dc79d5bcc8a816603511d8387914ab
README.md	8.98 KB	944ba5c90edd80f55769df7fb13c43f902baba6d47fadbe7db4489c6a6d26436

VersionsEach time the pinned upstream commit or file list changed in this repo. Newest first.

Upstream commit	Date	Contents
f6a596eaf9 pinned now	2026-11-15 04:55 UTC	23 files · 95.2 GB

Use it

$ fw get bartowski/MegaBeam-Mistral-7B-512k-GGUF
# manifest sha256 76c86376466d… · 95.2 GB · 23 files
$ fw verify ./MegaBeam-Mistral-7B-512k-GGUF bartowski/MegaBeam-Mistral-7B-512k-GGUF

fw.pyChecks the signature and every sha256, tries each source in order, resumes.

$ python3 fw.py get https://freeweights.allternit.com/manifests/bartowski/MegaBeam-Mistral-7B-512k-GGUF.json --dir ./MegaBeam-Mistral-7B-512k-GGUF
$ python3 fw.py verify ./MegaBeam-Mistral-7B-512k-GGUF ./bartowski__MegaBeam-Mistral-7B-512k-GGUF.json

hfThe Hugging Face CLI with one setting. Same repo id.

$ HF_ENDPOINT=https://freeweights.allternit.com hf download bartowski/MegaBeam-Mistral-7B-512k-GGUF --revision f6a596eaf9072a24f79a9b9a175d2ce103ca665b --local-dir ./MegaBeam-Mistral-7B-512k-GGUF

llama.cppDownload one GGUF variant (MegaBeam-Mistral-7B-512k-IQ2_M, 2.50 GB), then serve it.

$ HF_ENDPOINT=https://freeweights.allternit.com hf download bartowski/MegaBeam-Mistral-7B-512k-GGUF --include "MegaBeam-Mistral-7B-512k-IQ2_M*" --local-dir ./MegaBeam-Mistral-7B-512k-GGUF
$ llama-server -m ./MegaBeam-Mistral-7B-512k-GGUF/MegaBeam-Mistral-7B-512k-IQ2_M.gguf

OllamaMake an Ollama model from the downloaded file.

$ printf 'FROM ./MegaBeam-Mistral-7B-512k-GGUF/MegaBeam-Mistral-7B-512k-IQ2_M.gguf\n' > Modelfile
$ ollama create megabeam-mistral-7b-512k-gguf -f Modelfile

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: 944ba5c90edd80f55769df7fb13c43f902baba6d47fadbe7db4489c6a6d26436
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: bartowski <huggingface.co/bartowski>Newsgroups: alt.freeweights.mistralSubject: MegaBeam-Mistral-7B-512k-GGUFDate: 2026-11-15Commit: f6a596eSigned: 2DC28CC7B26A283B ✓

Upstream: bartowski/MegaBeam-Mistral-7B-512k-GGUF
Pinned commit: f6a596eaf9072a24f79a9b9a175d2ce103ca665b
Manifest: MegaBeam-Mistral-7B-512k-GGUF.json · .minisig
Manifest sha256: 76c86376466da9a8fde033300f572be4da3a30df361d16d76ca4f7765f3953b3
Signature: minisign, Ed25519, key 2DC28CC7B26A283B
Created: 2026-11-15 04:55 UTC
Generator: freeweights-manifest/0.1.0
Takedowns: None.

How to verify

$ curl -O https://freeweights.allternit.com/manifests/bartowski/MegaBeam-Mistral-7B-512k-GGUF.json
$ curl -O https://freeweights.allternit.com/manifests/bartowski/MegaBeam-Mistral-7B-512k-GGUF.json.minisig
$ minisign -Vm MegaBeam-Mistral-7B-512k-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

asp-prototyping/MegaBeam-Mistral-7B-512k · this model is a quantization of it

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

Get it

Download

$ HF_ENDPOINT=https://freeweights.allternit.com hf download bartowski/MegaBeam-Mistral-7B-512k-GGUF

Open in Freeweights Studio

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

Signed manifest

minisign, key 2DC28CC7B26A283B. MegaBeam-Mistral-7B-512k-GGUF.json. Provenance. Takedown history: none.

Source

huggingface.co/bartowski/MegaBeam-Mistral-7B-512k-GGUF

Related models

All of alt.freeweights.mistral →

From: bartowski <huggingface.co/bartowski>Subject: alamios_Mistral-Small-3.1-DRAFT-0.5B-GGUF

From:

microsoft <huggingface.co/microsoft>Subject: chatbench-mistral-7b

From: NousResearch <huggingface.co/NousResearch>Subject: DeepHermes-3-Mistral-24B-Preview

From:

mistralai <huggingface.co/mistralai>Subject: Devstral-Small-2-24B-Instruct-2512

From:

mlx-community <huggingface.co/mlx-community>Subject: Devstral-Small-2-24B-Instruct-2512-4bit

From:

mlx-community <huggingface.co/mlx-community>Subject: Devstral-Small-2-24B-Instruct-2512-6bit