CatalogWantedMirrorsLedgerOperating StatementDocsStudio statusCard system
alt.freeweights.text / tasktrove-dq-unix-step10-30b-a3b
From: laion <huggingface.co/laion>Newsgroups: alt.freeweights.textSubject:

tasktrove-dq-unix-step10-30b-a3b

Date: 2026-11-15Commit: 68ee41cSigned: 2DC28CC7B26A283B ✓

Reinforcement learning model with 30.5B parameters. Safetensors weights, 61.1 GB in all. Published by laion under Apache-2.0.

30.5B params61.1GBtasktrove-dq-unix-step10-30b-a3bsafetensorspermissive
FactsParams30.5BSize61.1 GBFormatssafetensorsTaskreinforcement learningLicenseApache-2.0From sourcepermissive
$ fw get laion/tasktrove-dq-unix-step10-30b-a3b
# manifest sha256 65d7cfff1832… · 61.1 GB · 33 files
$ fw verify ./tasktrove-dq-unix-step10-30b-a3b laion/tasktrove-dq-unix-step10-30b-a3b
From sourceHosted (signed)laion/tasktrove-dq-unix-step10-30b-a3bLast check: 2026-11-15 04:19 UTC
Run at homeGet help

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

Fine-tune of Qwen3-Coder-30B-A3B-Instruct.

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
61.1 GB
Weights
61.1 GB (safetensors)
Parameters
30.5B
Formats
safetensors
Task
reinforcement learning
License
Apache-2.0 (permissive)
Files
33, each with a sha256
Publisher
laion
Added
2026-11-15
Section
alt.freeweights.text
Lineage

Original release from the publisher.

Base model: Qwen/Qwen3-Coder-30B-A3B-Instruct

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 04:19 UTCpassed

Model card

From the publisher at commit 68ee41c. Shown as plain text with links; images open at the publisher.

tasktrove-dq-unix (step 10)#

RL checkpoint from the TaskTrove data-quality sweep, trained with SkyRL from Qwen/Qwen3-Coder-30B-A3B-Instruct using RLOO over agentic software-engineering tasks executed by OpenCode in sandboxed environments.

  • Base model: Qwen/Qwen3-Coder-30B-A3B-Instruct (Qwen3 MoE, 48 layers)
  • Checkpoint: global_step_10
  • Source run: rl-tasktrove-dq-sweep-30b-qwen3-coder-30-20260727-082204-e42f1d
  • Weights: 16 safetensors shards, 61.1 GB

What this is for#

The sweep measures dataset quality, not model quality. Each arm trains the same base model on a different TaskTrove source so the sources can be compared. These checkpoints are research artifacts for that comparison. None has been evaluated as a general-purpose model, and no benchmark numbers are claimed here.

Training configuration#

RLOO (advantage_estimator: rloo_n) with megatron backend, tensor-parallel 4, pipeline-parallel 2, expert-parallel 4, across 32 H100s. The objective is deliberately unregularized: use_kl_loss: false, use_entropy_loss: false, and policy_update_steps: 1, which leaves the PPO clip ratio inert at 0.0. That choice makes entropy dynamics the primary failure mode across the sweep, and it is why several arms ended early.

Provenance#

Exported from a torch.distributed.checkpoint megatron checkpoint by re-running the trainer's own export path (bridge.save_hf_weights) at the checkpoint's own step, so no offline conversion was involved. Shard count, index total_size and weight_map completeness were verified against the object store before upload.

Training Traces#

Training-time OpenCode/Harbor rollouts for this run are published as a companion dataset: laion/terminal_bench_2_tasktrove_dq_unix_step10_30b_a3b_20260730_014756

The dataset contains the last episode of each trial (per make_and_upload_trace_dataset --episodes last) — the rollouts the policy was trained on.

Training Logs#

training_logs/ holds the parsed metric surface for this run — per-step training metrics, vLLM engine metrics, a summary report, and the reward-vs-steps plot — alongside the raw trainer log. Capability tokens have been redacted from the logs.

Files

33 files · 61.1 GBWeights 61.1 GB (safetensors). Sizes and sha256 from the signed manifest.
Show all 33 files with their sha256
FileSizesha256
.gitattributes1.71 KB2e450ae81b5004310b3bb9f5f4e9c4362fd88d75f7a3cdc1303e7c6e26572d38
README.md2.50 KB71cdeff771c3ff84f0bd0dcc60207e777968aa2b0fa7771816e3cb61546d079b
chat_template.jinja6.21 KB5a38bfa05833266240066aedc497decc9b00cc0d3e3b8cceea98cf530196ab06
config.json1.02 KB3580b33db4d9c872e3d297884e3a07509a136f5399e878f53259cddb2c4f0eb6
generation_config.json216 Bf2a80a84a116bc7a296bbddfea4e48cf69a12f46e1925fe912b59c21a3651a90
model-00001-of-00016.safetensors4.00 GB6eb4b9a5cc937447d64ed7fe77a46ec9c2adf36b24374d5f34c23857adbd8107
model-00002-of-00016.safetensors4.00 GBae6bde9082bdc2ee799c93b1f0165981fc34fef5ce82f53ac46e3cd57194cc9e
model-00003-of-00016.safetensors4.00 GB6669c11d6bfc4f4273c2ee7e48eeaeeaa566738c4c5fdbc42d38add7c641b31e
model-00004-of-00016.safetensors4.00 GB5c08f5a0695af488a8c6a33146c136c9202b01e0ce7bf9d275138dc9e1c8fbec
model-00005-of-00016.safetensors4.00 GB7bffa31b9110a1290dccab653306c4d2eff095e97b878ea4e126d775fc746345
model-00006-of-00016.safetensors4.00 GB2e32fef47b5c8e112c7e8ee019681e863e72a67b7c8d0654ee9edfe3e4b04395
model-00007-of-00016.safetensors4.00 GBed69afe7c10601ec3b51e4f114dbc8f83fe6001909d91169eb87182a9e6c034a
model-00008-of-00016.safetensors4.00 GB151ac7faa4a8ddd8d1bca6ba0a5442f0b5694ba66039d3cfd392aeef2c714d02
model-00009-of-00016.safetensors4.00 GB243839e82aab8144f9d17052efbd5fcbbabf43805a061482e34109b2109c571b
model-00010-of-00016.safetensors4.00 GB61371ce2f8692dc0c5f2f65a0252a8a9477859896f03271b6548aa946cefb2d0
model-00011-of-00016.safetensors4.00 GB3c1f368907ec0958a48eccf8876526148bf750d450d40bf025d1bcae65362fee
model-00012-of-00016.safetensors3.99 GB7e95aa5c50d37acb2d8a93439548c1c5f098eb6cb101c1212c072d805581aee9
model-00013-of-00016.safetensors4.00 GB15e30e2a63a73bf58ad1d00ae7b45f36a1271a4bd76108a0081023c36c26d85f
model-00014-of-00016.safetensors4.00 GB8803e89807f58896bff49d5cf20e743fb63ff1cd3d809ffbe89d780b2e78f65e
model-00015-of-00016.safetensors4.00 GB9be1a6062dedf7820212f302530931fc7e81cc9b19adae18cd298f90e48092fe
model-00016-of-00016.safetensors1.09 GBd2126947b94b7da9978e9ac4f5591a7a1914429f8bb38093752d4984068793c4
model.safetensors.index.json1.78 MB4bb7fbce5a056ef9a3ae75ec33a5424fb12204bce319a09a6198b9306813c5e0
tokenizer.json11.4 MBbe75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
tokenizer_config.json695 Bf6f4c67578d26ccdf6535fa0bc88683b9a0e7acb6797aca1f4759ad3330c22e9
training_logs/20260730_095743_metrics_finelog.csv15.5 KB0460ff85391ea208d74892a13cb7bf2db0cc98dbf983c5282c66fd495352ada0
training_logs/20260730_095743_metrics_table.csv15.7 KBd0d360e4afde507a94fbd01c60e04de614ef7d6ead0bd853ee047cbfc6a954fd
training_logs/20260730_095743_vllm_metrics_finelog.csv122 KBa98a6b9a99eb943d1ed63a1fc3af77ff9ac89789296552ea534a6e94736848bb
training_logs/20260730_095743_vllm_metrics_table.csv134 KBa391585c7f7cd18de2a51eb09aa626da9306df2727d7c36bbb77249cbc64c664
training_logs/e42f1d_trainer.log18.4 MB3b200200b5bd44084ba1d008df22d2a17c56213329a14e85963ceeaee2e4c349
training_logs/metrics.csv15.7 KBd0d360e4afde507a94fbd01c60e04de614ef7d6ead0bd853ee047cbfc6a954fd
training_logs/report.md11.2 KBbd7afbfdc28c9c0d226fe543166764110a0ec8efedc7c69b6908f77f3c3a5a1d
training_logs/reward_plot.png139 KB819f369a1d6bf8f61c8a7e4c32bf8601d70c7158eb1520135008528a8d5b37dd
training_logs/vllm_metrics.csv134 KBa391585c7f7cd18de2a51eb09aa626da9306df2727d7c36bbb77249cbc64c664
VersionsEach time the pinned upstream commit or file list changed in this repo. Newest first.
Upstream commitDateContents
68ee41c6c2 pinned now2026-11-15 04:19 UTC33 files · 61.1 GB

Use it

$ fw get laion/tasktrove-dq-unix-step10-30b-a3b
# manifest sha256 65d7cfff1832… · 61.1 GB · 33 files
$ fw verify ./tasktrove-dq-unix-step10-30b-a3b laion/tasktrove-dq-unix-step10-30b-a3b
fw.pyChecks the signature and every sha256, tries each source in order, resumes.
$ python3 fw.py get https://freeweights.allternit.com/manifests/laion/tasktrove-dq-unix-step10-30b-a3b.json --dir ./tasktrove-dq-unix-step10-30b-a3b $ python3 fw.py verify ./tasktrove-dq-unix-step10-30b-a3b ./laion__tasktrove-dq-unix-step10-30b-a3b.json
hfThe Hugging Face CLI with one setting. Same repo id.
$ HF_ENDPOINT=https://freeweights.allternit.com hf download laion/tasktrove-dq-unix-step10-30b-a3b --revision 68ee41c6c28103862388c625ff7d1eb2d5a777d3 --local-dir ./tasktrove-dq-unix-step10-30b-a3b
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
71cdeff771c3ff84f0bd0dcc60207e777968aa2b0fa7771816e3cb61546d079b
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: laion <huggingface.co/laion>Newsgroups: alt.freeweights.textSubject: tasktrove-dq-unix-step10-30b-a3bDate: 2026-11-15Commit: 68ee41cSigned: 2DC28CC7B26A283B ✓
Upstream
laion/tasktrove-dq-unix-step10-30b-a3b
Pinned commit
68ee41c6c28103862388c625ff7d1eb2d5a777d3
Manifest
tasktrove-dq-unix-step10-30b-a3b.json · .minisig
Manifest sha256
65d7cfff1832aedd560eb5af70ed323153267799f4af147bd7a0800506baf384
Signature
minisign, Ed25519, key 2DC28CC7B26A283B
Created
2026-11-15 04:19 UTC
Generator
freeweights-manifest/0.1.0
Takedowns
None.

How to verify

$ curl -O https://freeweights.allternit.com/manifests/laion/tasktrove-dq-unix-step10-30b-a3b.json $ curl -O https://freeweights.allternit.com/manifests/laion/tasktrove-dq-unix-step10-30b-a3b.json.minisig $ minisign -Vm tasktrove-dq-unix-step10-30b-a3b.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-Coder-30B-A3B-Instruct · 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 laion/tasktrove-dq-unix-step10-30b-a3b
Open in Freeweights Studio

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

Signed manifest
minisign, key 2DC28CC7B26A283B. tasktrove-dq-unix-step10-30b-a3b.json. Provenance. Takedown history: none.
Source
huggingface.co/laion/tasktrove-dq-unix-step10-30b-a3b

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