alt.freeweights.text / jtd-d3-60-30B

From: laion <huggingface.co/laion>Newsgroups: alt.freeweights.textSubject:

jtd-d3-60-30B

Date: 2026-11-15Commit: e2f3b2fSigned: 2DC28CC7B26A283B ✓

Reinforcement learning model with 30.5B parameters. Safetensors weights, 61.7 GB in all. Published by laion under Apache-2.0.

30.5B params61.7GBjtd-d3-60-30Bsafetensorspermissive

FactsParams30.5BSize61.7 GBFormatssafetensorsTaskreinforcement learningLicenseApache-2.0From sourcepermissive

$ fw get laion/jtd-d3-60-30B
# manifest sha256 b49be09fb830… · 61.7 GB · 76 files
$ fw verify ./jtd-d3-60-30B laion/jtd-d3-60-30B

From sourceHosted (signed)laion/jtd-d3-60-30BLast check: 2026-11-15 04:26 UTC

Run at home Get 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

Run8 Fine-tune7 Convert8 Scan2

Runs with

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

Transformers LocalAI tinygrad Triton Inference Server BentoML CTranslate2 ExLlamaV2 LitServe

Overview

Size: 61.7 GB
Weights: 61.1 GB (safetensors)
Parameters: 30.5B
Formats: safetensors
Task: reinforcement learning
License: Apache-2.0 (permissive)
Files: 76, 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

Badge	Host	Last check	Result
From source	huggingface.co	2026-11-15 04:26 UTC	passed

Model card

From the publisher at commit e2f3b2f. Shown as plain text with links; images open at the publisher.

laion/jtd-d3-60-30B#

Jupiter TaskTrove DAPO campaign, arm D3 — Section-3.4-only arm — D0 + overlong shaping (per-trajectory, l_max 12288 / l_cache 3072). Dataset: laion/codeforces-v2 (10,000 competitive-programming tasks; late steps trained on the TaskTrove codeforces-v3 snapshot after laion/codeforces-v2 was deleted). Base model Qwen/Qwen3-Coder-30B-A3B-Instruct; FSDP2 fully-async trainer, 16 policy GPUs (4 nodes x 4 GH200) + 4 vLLM engines (TP2); Harbor/Daytona sandboxed terminus-2 rollouts.

Checkpoint: step 60 of 80, selected by trailing-5 EMA of reward/avg_raw_reward over the arm's full step series (campaign ended early on platform degradation + data-quality grounds; final banked step 63).
Launch config: rl_config.yaml in this repo; metrics + per-link logs in training_logs/.
pass@8 reads 1.0 on the dynamic-sampling arms (d1/d2/d4) are post-filter by construction; use reward/avg_raw_reward series in training_logs/ for cross-arm comparison.

Training Traces#

Training-time Daytona/Harbor rollouts: open-athena/jtd-d3 — the last episode of each trial, conserved at a 25% subsample: every 4th trial dir of the arm's full trace store (30815 enumerated at upload; 155 shards, 17393 rows). Owner-instructed conservation quota for this campaign's cleanup, not the uploader default.

Files

76 files · 61.7 GBWeights 61.1 GB (safetensors). Sizes and sha256 from the signed manifest.

Show all 76 files with their sha256

File	Size	sha256
.gitattributes	2.22 KB	d276493b61dab8b95485bb8c8ce490589a5e130c8248b201429ca508ec343184
README.md	1.56 KB	5c1f4c38fc889aa3b86fa788e514fef91218c4c766f56714a86ac4f4719f9d03
chat_template.jinja	6.21 KB	5a38bfa05833266240066aedc497decc9b00cc0d3e3b8cceea98cf530196ab06
config.json	1.02 KB	e58feff12a5ed7fce42eb6ae6054502ba786843a515a2f6f4053adf5fb75f2bd
generation_config.json	217 B	2cb42f3bbca69b7dbf925354f69a3a8c5ef2184f33f8c2cd744d24aaa8e6ca69
model-00001-of-00002.safetensors	50.0 GB	644315e646dc85e5c5f747d84fb228de4ecd4c33f70b702dac5d2d24a6a4d9cf
model-00002-of-00002.safetensors	11.1 GB	93e48edd635ab4f3b9f2f4355dd92afcc550e2affe1bf120a13bb923b44d3f0c
model.safetensors.index.json	1.70 MB	8056f5e61a77bcc39322368320783ace12dfd0877cf642821620e33341d5d25e
rl_config.yaml	16.9 KB	46d5927ce5722723e1569729c53dd023f454f9e20dd52068420837407a3681d7
tokenizer.json	11.4 MB	be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
tokenizer_config.json	694 B	700f87d8a901e6213c9b339ae3a0ff3f1deb2bfc39f1f66c85de3669a566fa84
training_logs/20260901_010234_metrics_jtd-d3_3_jtd-d3_1457188.csv	8.10 KB	8cbc843ccf002d5309cc0a857beb07799c85c9769e74ccfc568ba222f5337854
training_logs/20260901_010234_metrics_jtd-d3_3_jtd-d3_1457189.csv	7.24 KB	3902485919a8a89925aa5d7b26f65a9ffd4dea3e8ba916763269cec59aa4100c
training_logs/20260901_010234_metrics_jtd-d3_3_jtd-d3_1457190.csv	9.25 KB	d4f6a07d271cfc325c667f13ad523d4670de71d523e08df3e5f2d402c6e3a3f7
training_logs/20260901_010234_metrics_jtd-d3_3_jtd-d3_1457191.csv	8.80 KB	02f2df2d41783cadbda8da4e497ea9c4f5c3fe27633b66d5211ae70d5bc49f82
training_logs/20260901_010234_metrics_jtd-d3_3_jtd-d3_1457192.csv	8.01 KB	3ecbda7f77c4c0f24b5e77acb3d08196ac631cc0197b60f6b35b4d2463806aac
training_logs/20260901_010234_metrics_jtd-d3_3_jtd-d3_1457193.csv	8.04 KB	a98cc0653f98c7ca76f610f74ade39dab96b45f8fb708144fe21a76417f50845
training_logs/20260901_010234_metrics_jtd-d3_3_jtd-d3_1464917.csv	8.82 KB	6308016475eba9d8c7437d53d5f2ab3755e9eeed6ab79f106fb7c38b532aff11
training_logs/20260901_010234_metrics_jtd-d3_3_jtd-d3_1464919.csv	4.91 KB	a1c22e1e408dcda2ee539dd080e29081be515ec4ae2fdc98da352716aeffddd2
training_logs/20260901_010234_metrics_jtd-d3_3_jtd-d3_1464920.csv	9.87 KB	9dab2785ca66d399b2b4f6b7833813548ea44932acb8d278e2427d5f7dd242cc
training_logs/20260901_010234_metrics_jtd-d3_3_jtd-d3_1467314.csv	7.54 KB	aeb36501410fea07e253bb2afffb972870f5d8153fa81c7b540e58d0855cd705
training_logs/20260901_010234_metrics_jtd-d3_3_jtd-d3_1467315.csv	6.11 KB	21fa833784745edad5ca684dd27fb869996f3b67f1390e84a2ad176752391e3c
training_logs/20260901_010234_metrics_jtd-d3_3_jtd-d3_1487650.csv	4.30 KB	ca7224149b136d7e28b55d8bd5ccc5b4bf89562bac6e59a47b0ac635c7ef94ee
training_logs/20260901_010234_metrics_jtd-d3_3_jtd-d3_1494368.csv	5.74 KB	f1f9b8a2da2cf2dff54bd0846316aa89b5993b3c3fff15fc6febc2e2bc894c10
training_logs/20260901_010234_metrics_jtd-d3_4_jtd-d3_1527847.csv	8.96 KB	5bfcf5d4044b7da894d45fafe594faf8865999e68885352c2eb765bbfde316a7
training_logs/20260901_010234_metrics_jtd-d3_4_jtd-d3_1527848.csv	8.21 KB	8db9622af522ce34bfdda59e69ba4964b7d5a223cc828a18bf14159edbb4f988
training_logs/20260901_010234_metrics_jtd-d3_4_jtd-d3_1527849.csv	4.91 KB	c437cbf0e4a71d6aa5735507a800b56141fbd1fce26a3ff5ee95292371b4fe34
training_logs/20260901_010234_metrics_jtd-d3_5_jtd-d3_1558849.csv	4.98 KB	b9dc532767e8e3554cd8c72b754136ae0d60523d997a7d2ac877da0168d2371a
training_logs/20260901_010234_metrics_report.md	40.2 KB	9e8269583a998801bcddbd51a430e605a14a8c94f54e8a8e6bf2e05fc63a3e72
training_logs/20260901_010234_metrics_table.csv	65.9 KB	e4b05a1a5fffab5875d40bbd43982cf7c0deabaea1405b8b85fbd7da9ed33075
training_logs/20260901_010234_reward_vs_steps.png	272 KB	38986b6a720b8ddd2948358e050dae7c5563df936c2b647141c02f16d0da696c
training_logs/20260901_010234_trial_results.csv	9.17 MB	4eb787f89b7bf5c348f86e7a0912f535bcf20c33b9fa53f880d9e9b2d1e82777
training_logs/20260901_010234_turn_count_distribution.png	147 KB	cf71b83116503de6beabf9c65fb1d24402dc13cff1b2659f7a28cef03995f625
training_logs/20260901_010234_vllm_metrics_jtd-d3_3_jtd-d3_1457188.csv	271 KB	f241721b5ac55ce215a4623ffa62f2bcd1bc25433052f6bcde4da007aecc705e
training_logs/20260901_010234_vllm_metrics_jtd-d3_3_jtd-d3_1457189.csv	164 KB	7def38bf258f31be6e21c9aa2ee61b722c6e310f9b8b657237a093ed2cd75d82
training_logs/20260901_010234_vllm_metrics_jtd-d3_3_jtd-d3_1457190.csv	212 KB	9700df6dd81fe8fb427b8d98ac48e8d1add431239f6ef6a7bcb7dd2db42974dc
training_logs/20260901_010234_vllm_metrics_jtd-d3_3_jtd-d3_1457191.csv	174 KB	cdbc50dd47c391c5bd8bb53a9bfe69127bb458c196c1c186ac6c2b627c0bfc90
training_logs/20260901_010234_vllm_metrics_jtd-d3_3_jtd-d3_1457192.csv	157 KB	c29d80542883524a32e9f413d2105fa57efb723e46c1a50e05c15465971a66f9
training_logs/20260901_010234_vllm_metrics_jtd-d3_3_jtd-d3_1457193.csv	154 KB	d264cbf37d6e768573e85da676b0d09385bf29224e5aecd4456afbf76e7a3841
training_logs/20260901_010234_vllm_metrics_jtd-d3_3_jtd-d3_1464917.csv	190 KB	af78382d77fa7dd8e83ff2a84d5437ca420678a5059fb85284806374f2d13f0c
training_logs/20260901_010234_vllm_metrics_jtd-d3_3_jtd-d3_1464919.csv	118 KB	fc8d311eb172def119344975e52dd65739e40473de4a17ab71e59de63fccc816
training_logs/20260901_010234_vllm_metrics_jtd-d3_3_jtd-d3_1464920.csv	395 KB	8eb0620029c9f12be3a7678a9e5ee0c1eec33c1dd3dfb8b21e720a83400b59e3
training_logs/20260901_010234_vllm_metrics_jtd-d3_3_jtd-d3_1467314.csv	414 KB	e6e78ed387872e747029be53f3b744e6d6e0530bcd9ac405e665170e00f9d29b
training_logs/20260901_010234_vllm_metrics_jtd-d3_3_jtd-d3_1467315.csv	321 KB	414b425d7478fa59294bdcbef91926e3f2635026f738229befbb0f8167abe22c
training_logs/20260901_010234_vllm_metrics_jtd-d3_3_jtd-d3_1467316.csv	299 KB	3f9ebda5c9309f96845c8ddecd538b5adfacd6b0013d7d0957e09841c4aade14
training_logs/20260901_010234_vllm_metrics_jtd-d3_3_jtd-d3_1487649.csv	324 KB	161863ed7621e52ea203f3d562b21267b70ad0018dc5e70ffd29901a6c69c66e
training_logs/20260901_010234_vllm_metrics_jtd-d3_3_jtd-d3_1487650.csv	334 KB	d80c9e3ca811a9b7cdc37ff0022f8fea116d34a7be54ed04bd7a7a85e06e69c1
training_logs/20260901_010234_vllm_metrics_jtd-d3_3_jtd-d3_1487651.csv	132 KB	8ec36a36e6fafdf707ac94c797d37c9d9a751571d630ef4eeb6528e97e75e8b3
training_logs/20260901_010234_vllm_metrics_jtd-d3_3_jtd-d3_1494368.csv	442 KB	4f48a8c9cea30022302899baf1416b19c659cafe2cd21d00b06a4f5ad40b3e9e
training_logs/20260901_010234_vllm_metrics_jtd-d3_4_jtd-d3_1527847.csv	564 KB	b15cde4808cec090b82ba9d5d4a57bb5ec56cd4c831b5816d4cf3f890deffc22
training_logs/20260901_010234_vllm_metrics_jtd-d3_4_jtd-d3_1527848.csv	655 KB	072662b2773b8e06ccf88216f51acc83a5ea39b8af48d439dad658ac14dbe261
training_logs/20260901_010234_vllm_metrics_jtd-d3_4_jtd-d3_1527849.csv	658 KB	440ed44a0dcfe60043ca325fe7c53fd782f4195d987fef55cd7863a85bf26f49
training_logs/20260901_010234_vllm_metrics_jtd-d3_4_jtd-d3_1527850.csv	54.8 KB	2da942e60fdedc79cd41cb5802707d1aa671694a44b951f5dd1972b6a1cc953e
training_logs/20260901_010234_vllm_metrics_jtd-d3_5_jtd-d3_1558849.csv	582 KB	905fd0bb49a40818634bcc4fdbb3003b4b229c8d127a66b4986d3692edf99223
training_logs/20260901_010234_vllm_metrics_table.csv	8.50 MB	c71ff48196b363361d3be5c2bc5fe6b1060b1ee7e9df0d30ab1c7aa96444c15e
training_logs/jtd-d3_3_jtd-d3_1457188.out	3.94 MB	070ffe55aa7e277f5518db4dd71de764c1482782aadb77e91c0253620bf70a6a
training_logs/jtd-d3_3_jtd-d3_1457189.out	3.06 MB	031a4f04aa8e53d06d7c559da1e3701ae4120ca249ce53271e4f56e25a141687
training_logs/jtd-d3_3_jtd-d3_1457190.out	3.46 MB	612f94513b7f35e286213bc98ff5844cb3305090ad6d1e7e5e453b2e0bc4a43a
training_logs/jtd-d3_3_jtd-d3_1457191.out	3.33 MB	d0ecd4326b76b8ebc537f3ae6b43071b1ac3ee82f2f1d6519a245b1c69f6377d
training_logs/jtd-d3_3_jtd-d3_1457192.out	3.00 MB	4ba8562df3e68b4224f7e78bf23ee6fc4b78d86a556385620a20bc317458c110
training_logs/jtd-d3_3_jtd-d3_1457193.out	2.90 MB	a5a899e98d59cc2e6e770c7c83fcd6febe3ca3a408416b98e710d6ee8c4faaf0
training_logs/jtd-d3_3_jtd-d3_1464917.out	3.80 MB	651c827fa5b1339b6cafc78183f92b7c1eccb3d810ae3fe1e2ed6b326e9b053d
training_logs/jtd-d3_3_jtd-d3_1464919.out	2.10 MB	398df94ef6a52618a0c30a42d773808dff95cb5cfe2b11192587a5fdea98a91b
training_logs/jtd-d3_3_jtd-d3_1464920.out	5.96 MB	73cdcff0a305e5d837a63fdeacce1366b6554d41c43b0acb379f1313f879b7cd
training_logs/jtd-d3_3_jtd-d3_1467314.out	4.38 MB	c373e1f93d72d57d41a8165e0f8cd00a3ec8960c1f11cdaee7b25d1d65ab8b34
training_logs/jtd-d3_3_jtd-d3_1467315.out	3.36 MB	3d8c4d9dec82af281dd2d152498295b44cde1b8b93fe5fb6218c518e445a84dc
training_logs/jtd-d3_3_jtd-d3_1467316.out	71.4 MB	defe93a40c5a533407fdca7aa7180c3ff16b74c523c5234bd0abee51f47217cc
training_logs/jtd-d3_3_jtd-d3_1487649.out	41.4 MB	1b70b47213c78b929e7516b7589cbe81ad99769fc05f4502714d9cbae2b62124
training_logs/jtd-d3_3_jtd-d3_1487650.out	3.23 MB	f94a91ef9b6713d1bf165c72af165110b93ed9a253c07c16ef112f7bd46e0d19
training_logs/jtd-d3_3_jtd-d3_1487651.out	1.24 MB	dce7cf5f8c49243091bc9b05c7ec722f21bfec501cbafe8a98cac2e4e108e054
training_logs/jtd-d3_3_jtd-d3_1494368.out	280 MB	9b20c57869b51c6822fb658e148a9e020924ce1357d220cf6c10a2074e83dfec
training_logs/jtd-d3_4_jtd-d3_1527847.out	5.58 MB	6fc0f8d0e8398766086cfc68b38c22f01a90b15bc5064196793a03c46b21c692
training_logs/jtd-d3_4_jtd-d3_1527848.out	39.8 MB	a54c93b8286b43fd61032f11b44915fe684b508ed64bc469970c8eb15cafb408
training_logs/jtd-d3_4_jtd-d3_1527849.out	38.9 MB	cc68c26eea8c8b9198ec3e0f77fdd83164e1a53643d922fad81f758566da27e9
training_logs/jtd-d3_4_jtd-d3_1527850.out	2.08 MB	2904449472fd4f100a96440ad5f78932418ddf9456c3f25c5357600ba98ef238
training_logs/jtd-d3_5_jtd-d3_1558849.out	35.5 MB	2cdbff84dae39b2f6baa4c8f54a18f9bea668295803a606ae467556e5767f123

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

Upstream commit	Date	Contents
e2f3b2f278 pinned now	2026-11-15 04:26 UTC	76 files · 61.7 GB

Use it

$ fw get laion/jtd-d3-60-30B
# manifest sha256 b49be09fb830… · 61.7 GB · 76 files
$ fw verify ./jtd-d3-60-30B laion/jtd-d3-60-30B

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

$ python3 fw.py get https://freeweights.allternit.com/manifests/laion/jtd-d3-60-30B.json --dir ./jtd-d3-60-30B
$ python3 fw.py verify ./jtd-d3-60-30B ./laion__jtd-d3-60-30B.json

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

$ HF_ENDPOINT=https://freeweights.allternit.com hf download laion/jtd-d3-60-30B --revision e2f3b2f278b9eb6fb4aed173444b548b06631378 --local-dir ./jtd-d3-60-30B

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: 5c1f4c38fc889aa3b86fa788e514fef91218c4c766f56714a86ac4f4719f9d03
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: jtd-d3-60-30BDate: 2026-11-15Commit: e2f3b2fSigned: 2DC28CC7B26A283B ✓

Upstream: laion/jtd-d3-60-30B
Pinned commit: e2f3b2f278b9eb6fb4aed173444b548b06631378
Manifest: jtd-d3-60-30B.json · .minisig
Manifest sha256: b49be09fb830b1cd3d99505daf891635bcada662bbae5d14b5e25ecb16bf138b
Signature: minisign, Ed25519, key 2DC28CC7B26A283B
Created: 2026-11-15 04:26 UTC
Generator: freeweights-manifest/0.1.0
Takedowns: None.

How to verify

$ curl -O https://freeweights.allternit.com/manifests/laion/jtd-d3-60-30B.json
$ curl -O https://freeweights.allternit.com/manifests/laion/jtd-d3-60-30B.json.minisig
$ minisign -Vm jtd-d3-60-30B.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/jtd-d3-60-30B

Open in Freeweights Studio

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

Signed manifest

minisign, key 2DC28CC7B26A283B. jtd-d3-60-30B.json. Provenance. Takedown history: none.

Source

huggingface.co/laion/jtd-d3-60-30B

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