CatalogWantedMirrorsLedgerOperating StatementDocsStudio statusCard system
alt.freeweights.text / jtd-d2-20-30B
From: laion <huggingface.co/laion>Newsgroups: alt.freeweights.textSubject:

jtd-d2-20-30B

Date: 2026-11-15Commit: 426c163Signed: 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-d2-20-30Bsafetensorspermissive
FactsParams30.5BSize61.7 GBFormatssafetensorsTaskreinforcement learningLicenseApache-2.0From sourcepermissive
$ fw get laion/jtd-d2-20-30B
# manifest sha256 c67ef2c09496… · 61.7 GB · 51 files
$ fw verify ./jtd-d2-20-30B laion/jtd-d2-20-30B
From sourceHosted (signed)laion/jtd-d2-20-30BLast check: 2026-11-15 04:20 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
61.7 GB
Weights
61.1 GB (safetensors)
Parameters
30.5B
Formats
safetensors
Task
reinforcement learning
License
Apache-2.0 (permissive)
Files
51, 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 04:20 UTCpassed

Model card

From the publisher at commit 426c163. Shown as plain text with links; images open at the publisher.

laion/jtd-d2-20-30B#

Jupiter TaskTrove DAPO campaign, arm D2 — full-DAPO arm — D1 + DAPO §3.4 overlong shaping (per-trajectory, l_max 12288 / l_cache 3072), filter on unshaped rewards. Dataset: laion/codeforces-v2 (10,000 competitive-programming tasks; ckpt_interval 1). 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 20 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 20).
  • 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-d2 — the last episode of each trial, conserved at a 25% subsample: every 4th trial dir of the arm's full trace store (28495 enumerated at upload; 143 shards, 25433 rows). Owner-instructed conservation quota for this campaign's cleanup, not the uploader default.

Files

51 files · 61.7 GBWeights 61.1 GB (safetensors). Sizes and sha256 from the signed manifest.
Show all 51 files with their sha256
FileSizesha256
.gitattributes2.21 KB106a1d35567be3faa14abbdc19a053b7ceadc5fb77d32c1c526fdf91b129b921
README.md1.51 KB41eac8ee23d1360269db734262c136ff8a823b3ab2b06c7745cd75a846eaa277
chat_template.jinja6.21 KB5a38bfa05833266240066aedc497decc9b00cc0d3e3b8cceea98cf530196ab06
config.json1.02 KBe58feff12a5ed7fce42eb6ae6054502ba786843a515a2f6f4053adf5fb75f2bd
generation_config.json217 B2cb42f3bbca69b7dbf925354f69a3a8c5ef2184f33f8c2cd744d24aaa8e6ca69
model-00001-of-00002.safetensors50.0 GBfee7334b88ecbd1393b16695b2b82429dada0848dd25a38265a01047a2d1f35b
model-00002-of-00002.safetensors11.1 GB7f160f8f7ba3a9df65b8ba1eb1eb7cda0db07b9f33aaf412145dec2d21defde4
model.safetensors.index.json1.70 MB8056f5e61a77bcc39322368320783ace12dfd0877cf642821620e33341d5d25e
rl_config.yaml17.3 KB3e1f14cc25ac9bc4a1a668aff909ddcd4f189b7eda14231c34b104ee399e3973
tokenizer.json11.4 MBbe75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
tokenizer_config.json694 B700f87d8a901e6213c9b339ae3a0ff3f1deb2bfc39f1f66c85de3669a566fa84
training_logs/20260901_001504_metrics_jtd-d2_jtd-d2_1466109.csv8.33 KB0d0accf4be3c4eda41bd9a970537e5b078bef2a18e6c2ce1a95dcd9148343423
training_logs/20260901_001504_metrics_jtd-d2_jtd-d2_1466110.csv6.16 KB3f7feee561732cd9df99455715b7d7e10a64275864595327272e71ea23e2885b
training_logs/20260901_001504_metrics_jtd-d2_jtd-d2_1466111.csv5.32 KBb1b11654fdbe5ae358efa00282e7338ce27181ff2015ee6f470994bcbe423779
training_logs/20260901_001504_metrics_jtd-d2_jtd-d2_1487659.csv5.45 KB85659f7a35d28ae817f567ce47a2f9fe8c129959c5cb82acfd59a2ecbdbc578d
training_logs/20260901_001504_metrics_jtd-d2_jtd-d2_1506693.csv5.46 KB145ae5cc40936c6df371e5a879493967f7d8ddbd083c0143b4b3069f39172b33
training_logs/20260901_001504_metrics_jtd-d2_jtd-d2_1506694.csv6.25 KB153e709a48f61af2971dd9e96d07db8eb7bf2c8e141a6d3c2533b8c452a622a0
training_logs/20260901_001504_metrics_jtd-d2_jtd-d2_1506695.csv5.46 KB9bffd496483a0d0492903546727d0ada73b2b571dd99e7ca04fe2be779bfc530
training_logs/20260901_001504_metrics_jtd-d2_jtd-d2_1506696.csv4.83 KB6a3802231331c36fb67eae6c0b05c039e597e0163378a439da9f5d5194f46af4
training_logs/20260901_001504_metrics_report.md31.1 KB4ecc6360bfeabdc572a162601a379d9ebb65ccdb1bc2a13ea66bedf26efeb4ab
training_logs/20260901_001504_metrics_table.csv20.3 KB7cd1c4e7143ef35c103b9b2a5efa0b9d6294b06b76f45ee779423e5663208e11
training_logs/20260901_001504_reward_vs_steps.png189 KB8c8f41327dfc14639c630f3de827dd7e98d3efceab22847db801a77bbe89d10e
training_logs/20260901_001504_trial_results.csv8.91 MB2a3cab93b2d6d694d594d043d1de3ff958646af5b3e87821c88509268eb39c12
training_logs/20260901_001504_turn_count_distribution.png132 KB5830168cc5cda84e3c2ebd5ba8626a1aad1abf236ff18f7f3d3267e1263ed0f7
training_logs/20260901_001504_vllm_metrics_jtd-d2_jtd-d2_1466109.csv686 KB1c0e97e8dadbfb094c595779a31129425d08f7f3ea0dd23f4ecaabf541c69fec
training_logs/20260901_001504_vllm_metrics_jtd-d2_jtd-d2_1466110.csv678 KB33a31eb3c9105f2bb606d00a0c01aca742b08fe0610f4775530cb7eaf99ced44
training_logs/20260901_001504_vllm_metrics_jtd-d2_jtd-d2_1487647.csv212 KB0629c7e586c2005ae5b26c4578eab859d15390340a853d0b6514abfe9a6252f8
training_logs/20260901_001504_vllm_metrics_jtd-d2_jtd-d2_1487659.csv676 KB38eecd99b281dc6b38e15674605da639c4065efeec4178d5fa448eb02f66562b
training_logs/20260901_001504_vllm_metrics_jtd-d2_jtd-d2_1506693.csv668 KBb97ff8652f622fe561183dd3beac05442cebe1194ffa8575b20b568417c2e11b
training_logs/20260901_001504_vllm_metrics_jtd-d2_jtd-d2_1506694.csv618 KB7b1e56de2df94f1619e166a5f9f269a42e07d567a2b6c0a17ed9d904f5eca25a
training_logs/20260901_001504_vllm_metrics_jtd-d2_jtd-d2_1506695.csv581 KB61a17ebdb3d5b5ade9e483a4dfcbd49b81f16c58b175b49ab9e81e824fcbd598
training_logs/20260901_001504_vllm_metrics_jtd-d2_jtd-d2_1506696.csv658 KB2966a0a22e5fc048c2c8858efcf71657b4adffdfca848c5d9f2eb90f7aef25de
training_logs/20260901_001504_vllm_metrics_jtd-d2_jtd-d2_1506697.csv654 KB7fbb6cabf0ea6faabe3aa032b6670f551dc1bc82cbcc4e51b1a2060f36a0aba8
training_logs/20260901_001504_vllm_metrics_jtd-d2_jtd-d2_1527900.csv558 KB1be324905aae257c4d385c1192bc4cc1221a917564c7e72d6783d52257830a32
training_logs/20260901_001504_vllm_metrics_jtd-d2_jtd-d2_1527902.csv653 KB91a092df750a693a7db34b034b8c14eacd196c53fd25b04de1d353a0ea98df19
training_logs/20260901_001504_vllm_metrics_table.csv8.35 MB2cbf6e6893f6a23657b488eb20428b43062dd85a12aeb522a322d5ca1dc3a0af
training_logs/jtd-d2_jtd-d2_1466109.out6.79 MB10ccb86f1b68bf3ed5d38a05b901eee6b679b9ee301c1e1c417ef466cb513c3d
training_logs/jtd-d2_jtd-d2_1466110.out6.38 MBaefafaf1acb1d0e947fe2bf955dd0945d22f4740734c738dfc8834cc39dab1e0
training_logs/jtd-d2_jtd-d2_1466111.out63.6 MBcb9e8cc824642c11a48c91ebd55e477aa350b2204e842acd8a761dacdca865ad
training_logs/jtd-d2_jtd-d2_1487647.out37.5 MB34e562f1aa9d0e5355c0217e4e241953199d740ac3b0361902569d99ae38be34
training_logs/jtd-d2_jtd-d2_1487659.out6.42 MB6b0080851c1af19990a6200dc73a54bb8a83a3b3419ebf01cffefbf9a81fd9cc
training_logs/jtd-d2_jtd-d2_1487660.out331 MBaae14c95e298e445ee4c1e5c5952d3679e61886dfb04d1f4e90ee53b6cf4f25a
training_logs/jtd-d2_jtd-d2_1506693.out7.10 MB5dd500da8ce26e5915b56d647b8e3afeabd052cca0277eeef0b3aa195789c423
training_logs/jtd-d2_jtd-d2_1506694.out7.58 MBb850c355191982718ac4d5dfaff08bb1b9b91f0847ac67055428a685a0607198
training_logs/jtd-d2_jtd-d2_1506695.out5.95 MBccdeacc9740ff396105a849d9b87569d14cdf6f8cc99b7ffacf53625ff9e99f5
training_logs/jtd-d2_jtd-d2_1506696.out6.30 MB8d686c7185d26ee63bd3836fe3bb46e8a68971e018e1f766b014066b3cdaf419
training_logs/jtd-d2_jtd-d2_1506697.out46.9 MB91701c82c1a7f3410377ad071530c73b4161c123083ba2874af99c7d82757323
training_logs/jtd-d2_jtd-d2_1527900.out36.5 MB53be44ddd4090436f7914a2d900f22bba78965d7a10d0a5e166c0e12417049de
training_logs/jtd-d2_jtd-d2_1527901.out172 KBa5d97fbafd087d3dd24583e1d464b701907144aa33ea7cce930a9357508c8084
training_logs/jtd-d2_jtd-d2_1527902.out47.4 MB2348230d1905bc87fecc483cb7d608639f0e134e048ef741d5d3605be04c01af
training_logs/jtd-d2_jtd-d2_1536986.out26.1 KB5f6ab82c8083785553e3debf2a59393d2066627a21865f567c94183c7598a0d8
VersionsEach time the pinned upstream commit or file list changed in this repo. Newest first.
Upstream commitDateContents
426c163814 pinned now2026-11-15 04:20 UTC51 files · 61.7 GB

Use it

$ fw get laion/jtd-d2-20-30B
# manifest sha256 c67ef2c09496… · 61.7 GB · 51 files
$ fw verify ./jtd-d2-20-30B laion/jtd-d2-20-30B
fw.pyChecks the signature and every sha256, tries each source in order, resumes.
$ python3 fw.py get https://freeweights.allternit.com/manifests/laion/jtd-d2-20-30B.json --dir ./jtd-d2-20-30B $ python3 fw.py verify ./jtd-d2-20-30B ./laion__jtd-d2-20-30B.json
hfThe Hugging Face CLI with one setting. Same repo id.
$ HF_ENDPOINT=https://freeweights.allternit.com hf download laion/jtd-d2-20-30B --revision 426c163814d9988eff0159727d5e75ca2e339074 --local-dir ./jtd-d2-20-30B
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
41eac8ee23d1360269db734262c136ff8a823b3ab2b06c7745cd75a846eaa277
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-d2-20-30BDate: 2026-11-15Commit: 426c163Signed: 2DC28CC7B26A283B ✓
Upstream
laion/jtd-d2-20-30B
Pinned commit
426c163814d9988eff0159727d5e75ca2e339074
Manifest
jtd-d2-20-30B.json · .minisig
Manifest sha256
c67ef2c094961c2a5defaa358a7ddc1b026187f9999c238b2a677888a7b57cce
Signature
minisign, Ed25519, key 2DC28CC7B26A283B
Created
2026-11-15 04:20 UTC
Generator
freeweights-manifest/0.1.0
Takedowns
None.

How to verify

$ curl -O https://freeweights.allternit.com/manifests/laion/jtd-d2-20-30B.json $ curl -O https://freeweights.allternit.com/manifests/laion/jtd-d2-20-30B.json.minisig $ minisign -Vm jtd-d2-20-30B.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-d2-20-30B
Open in Freeweights Studio

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

Signed manifest
minisign, key 2DC28CC7B26A283B. jtd-d2-20-30B.json. Provenance. Takedown history: none.
Source
huggingface.co/laion/jtd-d2-20-30B

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