CatalogWantedMirrorsLedgerOperating StatementDocsStudio statusCard system
alt.freeweights.text / LAR-27B
From: amazon <huggingface.co/amazon>Newsgroups: alt.freeweights.textSubject:

LAR-27B

Date: 2026-11-15Commit: 468ca9dSigned: 2DC28CC7B26A283B ✓

Model with 26.9B parameters. Safetensors weights, 53.8 GB in all. Published by amazon under Apache-2.0.

26.9B params53.8GBLAR-27Bsafetensorspermissive
FactsParams26.9BSize53.8 GBFormatssafetensorsTasknot taggedLicenseApache-2.0From sourcepermissive
$ fw get amazon/LAR-27B
# manifest sha256 38b429a6c0d6… · 53.8 GB · 29 files
$ fw verify ./LAR-27B amazon/LAR-27B
From sourceHosted (signed)amazon/LAR-27BLast check: 2026-11-15 03:53 UTC
Run at homeGet help

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

Fine-tune of Qwen3.8-27B.

Works with

Run8Fine-tune7Convert8Scan2

Runs with

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

TransformersLocalAItinygradTriton Inference ServerBentoMLCTranslate2ExLlamaV2LitServe
OverviewModel cardFilesUse itLicenseProvenanceModel tree

Overview

Size
53.8 GB
Weights
53.8 GB (safetensors)
Parameters
26.9B
Formats
safetensors
Task
not tagged
License
Apache-2.0 (permissive)
Files
29, each with a sha256
Publisher
amazon
Added
2026-11-15
Section
alt.freeweights.text
Lineage

Original release from the publisher.

Base model: Qwen/Qwen3.8-27B

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 03:53 UTCpassed

Model card

From the publisher at commit 468ca9d. Shown as plain text with links; images open at the publisher.

LAR-27B#

This is the official model repository for Agentic RL Model with Process Reward with Lean 4 (LAR).

Model performance#

Main Results#

Image: ResultPlot

ModelSizeSWE-Bench-VerifiedTB-2.0TB-2.1
LAR-9B9B64.2%32.8%29.5%
LAR-27B27B84.2%68.2%77.3%
1. Install dependencies#
1.1 Install environment#
# Install the corresponding environment
# (run from the repository root; needs python3.12 + venv, git, docker with compose v2 (user in the docker group), sudo,
#  an NVIDIA driver and CUDA toolkit for CUDA 13; MODEL = local LAR-27B checkpoint, WORK = work dir on a fast local disk)
export MODEL=amazon/LAR-27B WORK=/path/to/local/disk/tb21
python3.12 -m venv ~/.venvs/lar-serve
~/.venvs/lar-serve/bin/pip install -r PSWF/EvaluationKit/TB-2.1/requirements/serve.txt ./PSWF/EvaluationKit/TB-2.1
python3.12 -m venv ~/.venvs/lar-eval
~/.venvs/lar-eval/bin/pip install -r PSWF/EvaluationKit/TB-2.1/requirements/eval.txt ./PSWF/Harness/Terminus-2 ./PSWF/EvaluationKit/TB-2.1
~/.venvs/lar-eval/bin/python -m pswf_harbor_patches --tokenizer-dir $MODEL
1.2 Init TB-2.1 containers#
# Initialize the TB-2.1 containers
~/.venvs/lar-eval/bin/python -m tb21_eval.prepare --work-dir $WORK
sudo ~/.venvs/lar-eval/bin/python -m tb21_eval.netguard up --work-dir $WORK
~/.venvs/lar-eval/bin/python -m tb21_eval.netguard probe --work-dir $WORK
2. Run TB-2.1 evaluation using PSWF with Terminus-2#
# Run the evaluation
~/.venvs/lar-serve/bin/python -m tb21_eval.serve up --model $MODEL --work-dir $WORK
~/.venvs/lar-eval/bin/python -m tb21_eval.make_job --model $MODEL --work-dir $WORK --run-dir $WORK/runs/x5 --n-attempts 5 --n-concurrent 40
setsid nohup ~/.venvs/lar-eval/bin/harbor run -c $WORK/runs/x5/job.json --env-file $WORK/runs/x5/harbor.env > $WORK/runs/x5/harbor.log 2>&1 &
# ~16 h in the background; score at any time (if harbor was interrupted: ~/.venvs/lar-eval/bin/python -m tb21_eval.resume --run-dir $WORK/runs/x5)
~/.venvs/lar-eval/bin/python -m tb21_eval.score $WORK/runs/x5

Training Data and Methodology#

The released checkpoint is post-trained from Qwen/Qwen3.8-27B (licensed under Apache License 2.0) using Lean4AgenticRL process reward methods. This version was trained only on a filtered subset of SWE-Rebench containing software-engineering task examples from public code repositories. The underlying repository licenses were verified at the pinned base commits, and only examples meeting the license-filtering criteria were retained.

Responsible AI Considerations#

At Amazon, we are committed to developing AI responsibly and take a people-centric approach that prioritizes education, science, and our customers, to integrate responsible AI across the end-to-end AI lifecycle. We believe the use of AI must respect the rule of law and human rights, and we encourage the safe and responsible development of AI. When downloaded or used in accordance with AWS Responsible AI Policy, developers should work with their internal model team to ensure this model meets requirements for the relevant industry and use case and addresses unforeseen product misuse.

Please report model quality, risk, security vulnerabilities or Amazon AI Concerns here.

Reference#

If you find our work useful, please cite it using the BibTeX entry below. This entry currently links to the PDF hosted on GitHub and will be updated once the arXiv version is available.

@misc{wang2026lean4agenticrl,
      title={Lean4AgenticRL: Advancing Agentic RL with Verifiable Multi-level Process Rewards}, 
      author={Ruida Wang and Gerald Friedland and Tong Zhang and Chris Kong},
      year={2026},
      eprint={2610.22448},
      archivePrefix={arXiv},
      primaryClass={cs.LG},
      url={https://arxiv.org/abs/2610.22448}, 
}

Files

29 files · 53.8 GBWeights 53.8 GB (safetensors). Sizes and sha256 from the signed manifest.
Show all 29 files with their sha256
FileSizesha256
.eval_results/swe_bench_verified.yaml182 Ba215a3e89e895b79e8c96dc2d8275725548a989e2d37e09a8046b328a4e60748
.eval_results/terminal_bench_2_0.yaml183 B96ce67385a03265536381ac1ac05d75d5d26afc5816a6043ec44e44dfeb23099
.eval_results/terminal_bench_2_1.yaml185 B98f49d898944ec9379318073693b7431d765a2b3f114c1edc71d7f1544ac74d0
.gitattributes1.57 KB34448b82c17d60fec9b65b1f093c115ddbaadc04beb1b0140b6bfed2e012a930
LICENSE11.5 KBbbedc3fda3305820b977265f01b8619d87570a6739de3a5582c3464840f1e57a
README.md4.21 KBc513e16b04de4067ebad9e8f17746857ac13022d532d697b2daa47199ff00e93
assets/ResultPlot.png627 KBc8e749678f845091fb73dbf53930be14d5f87133beb94e03e7640b146ec1fa2f
chat_template.jinja8.95 KBc3cf9e34abf4f9e36c2d72165aa9c132d3e2a725b6c2586aaa3a8af9d7a81041
config.json4.31 KB191e0af232104ed8b65258cf3fb2b842e288008baca7633c11b82a1ac7203aab
crc32.txt238 Bb42dd291f5f98b05807e458f5b47849969a9b3326acbf0c97d28b05826740b83
generation_config.json202 Be70c136c1b78ddc1fb0905bac8e733a4dc448d4f852a5dd75143fffc70be550e
merges.txt3.35 MBa9d356d7bdf1ef4949e3e748e95b8e10ad9d4e2e838eddc38a0a7b6b94d1db8d
model-00000-of-00011.safetensors5.28 GB74e89645f79edbe43236794fc054d4f58b9dc618586460fc8aab5561114b9b19
model-00001-of-00011.safetensors3.82 GBdc2d499e7cfa2f80a599be57915c3ab2ccdb2d944e99967d45907c061a289a33
model-00002-of-00011.safetensors5.36 GB6c7729606e381809d8dfd8c7def1fad943fa8b64d57f8ae740af812a589a6302
model-00003-of-00011.safetensors5.29 GB92787b75c22c09d38cff9d657732d8b8f563a3a6ffc157827fe783408e199e79
model-00004-of-00011.safetensors5.35 GB158d271ee674f50595c72befd00e48ecca87eb154301bb1c4348c60ea4d205ce
model-00005-of-00011.safetensors5.21 GB4081233c82677640e56f51ee042ed735b7b1f04e79a2611b738f9db3edbfca16
model-00006-of-00011.safetensors5.35 GBdca52ae3655a56b6f425dfcf80c559e0077176d9f9ac220aac466d7e7ba111b7
model-00007-of-00011.safetensors5.23 GBc2628f90b7a03d732e6a4bb04b7c2229cd207203b88bacb90e64fe9c3824f235
model-00008-of-00011.safetensors4.36 GB8dd2367a0588fefef10d7844a4ee1ac51af43275959a26ff6ac92d1b9e52f7fb
model-00009-of-00011.safetensors5.32 GB5c88ae6f45c1e2e4ba2335070b3615093e6a38580340e1d0cbe3758d950cbc41
model-00010-of-00011.safetensors3.22 GB108788b15457437dd9ef8e58a7df06ddf04c7d17c01bc0fe636cacf57bc1f39f
model.safetensors.index.json83.9 KB3bf69b2f9ae0249c3bd24992e28a2669aab0159518fc0ff8e1947cd8299eecb1
preprocessor_config.json390 B27225450ac9c6529872ee1924fcb0962ff5634834f817040f444118116f4e516
tokenizer.json12.8 MB0997f410c57a1f4e53b09e4be8f4a172d90edd9564368fb0847030937229b9f3
tokenizer_config.json17.9 KBb11349aafa7cdc6a320767cf7ceb29ed82f7eda5d65e8e0819e76f0ce947bf27
video_preprocessor_config.json385 B7768af27c1fafa9cc9011c1dc20067e03f8915e03b63504550e11d5066986d13
vocab.json6.72 MBce99b4cb2983d118806ce0a8b777a35b093e2000a503ebde25853284c9dfa003
VersionsEach time the pinned upstream commit or file list changed in this repo. Newest first.
Upstream commitDateContents
468ca9d829 pinned now2026-11-15 03:53 UTC29 files · 53.8 GB

Use it

$ fw get amazon/LAR-27B
# manifest sha256 38b429a6c0d6… · 53.8 GB · 29 files
$ fw verify ./LAR-27B amazon/LAR-27B
fw.pyChecks the signature and every sha256, tries each source in order, resumes.
$ python3 fw.py get https://freeweights.allternit.com/manifests/amazon/LAR-27B.json --dir ./LAR-27B $ python3 fw.py verify ./LAR-27B ./amazon__LAR-27B.json
hfThe Hugging Face CLI with one setting. Same repo id.
$ HF_ENDPOINT=https://freeweights.allternit.com hf download amazon/LAR-27B --revision 468ca9d8298afc1506daeffafbb3b5405be7a45b --local-dir ./LAR-27B
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
LICENSE
Text sha256
bbedc3fda3305820b977265f01b8619d87570a6739de3a5582c3464840f1e57a
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: amazon <huggingface.co/amazon>Newsgroups: alt.freeweights.textSubject: LAR-27BDate: 2026-11-15Commit: 468ca9dSigned: 2DC28CC7B26A283B ✓
Upstream
amazon/LAR-27B
Pinned commit
468ca9d8298afc1506daeffafbb3b5405be7a45b
Manifest
LAR-27B.json · .minisig
Manifest sha256
38b429a6c0d67169098727ae007cdb6ca4558623b70d2cbab1ca27d97df30bb7
Signature
minisign, Ed25519, key 2DC28CC7B26A283B
Created
2026-11-15 03:53 UTC
Generator
freeweights-manifest/0.1.0
Takedowns
None.

How to verify

$ curl -O https://freeweights.allternit.com/manifests/amazon/LAR-27B.json $ curl -O https://freeweights.allternit.com/manifests/amazon/LAR-27B.json.minisig $ minisign -Vm LAR-27B.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/Qwen3.8-27B · this model is a finetune of it

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

Get it
Download
$ HF_ENDPOINT=https://freeweights.allternit.com hf download amazon/LAR-27B
Open in Freeweights Studio

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

Signed manifest
minisign, key 2DC28CC7B26A283B. LAR-27B.json. Provenance. Takedown history: none.
Source
huggingface.co/amazon/LAR-27B

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