Downloads · 30 days
33
38% of all-time downloads
jtd-d0-69-30B
laion/jtd-d0-69-30B
jtd-d0-69-30B is a reinforcement learning model from laion. Use it for the reinforcement learning task on the model card, and read the license before you ship it in a product. It is set up for transformers. The card lists the license as apache-2.0.
GRPO control checkpoint (step 69 of 80; the run ended early at step 72 after its restart chain exhausted) from the Jupiter TaskTrove DAPO campaign, arm D0 — plain GRPO, symmetric clipping (eps 0.2/0.2), token-level lo…
All-time downloads
86
Public
Parameters
30.5B
61.1 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors61.1 GB · 100%
At a glance
- Task
- Reinforcement Learning
- Library
- transformers
- License
- apache-2.0
- Model type
- qwen3_moe
- Access
- Public
- Created
- Aug 25, 2026
- Updated
- Sep 4, 2026
- SHA
- 6f7528d2
Base models
Model card
From the Hugging Face model README
laion/jtd-d0-69-30B
GRPO control checkpoint (step 69 of 80; the run ended early at step 72 after its restart chain exhausted) from the Jupiter TaskTrove DAPO campaign, arm D0 — plain GRPO, symmetric clipping (eps 0.2/0.2), token-level loss mean, no dynamic sampling, no reward shaping.
- Base model: Qwen/Qwen3-Coder-30B-A3B-Instruct
- Training data:
laion/codeforces-v2(10,000 competitive-programming tasks, sandboxed agent rollouts via Harbor/Daytona, terminus-2 harness, FSDP2 fully-async GRPO on 16× GH200) - Checkpoint selection: trailing-5 EMA of
reward/avg_raw_reward(0.2755 at step 69 vs 0.2354 at step 72) - Launch config:
rl_config.yamlin this repo - Metrics + logs:
training_logs/in this repo (per-link metrics CSVs, trial results, report)
Training Traces
Training-time Daytona/Harbor rollouts: open-athena/jtd-d0
(the last episode of each trial — the rollouts the policy trained on after rollback/truncation).
- Task
- Reinforcement Learning
- Library
- transformers
- Type
- qwen3_moe
- License
- apache-2.0
- Created
- Aug 25, 2026
- Updated
- Sep 4, 2026