Downloads · 30 days
21
33% of all-time downloads
Qwen3.6-35B-A3B-countdown-GRPO-20260909
semianalysisai/Qwen3.6-35B-A3B-countdown-GRPO-20260909
Qwen3.6-35B-A3B-countdown-GRPO-20260909 is a reinforcement learning model from semianalysisai. 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.
A specialist checkpoint for countdown puzzles, trained from Qwen3.6-35B-A3B with 40 updates of verifier-reward GRPO (Group Relative Policy Optimization).
All-time downloads
63
Public
Parameters
35.5B
71 GB on disk
Likes
1
Public
Repo makeup
Click a slice to open those files.
.safetensors71 GB · 100%
Parameter types
How the weights are stored.
BF1635.5B · 100%
Base models
Model card
From the Hugging Face model README
Qwen3.6-35B-A3B · Countdown
A specialist checkpoint for countdown puzzles, trained from Qwen3.6-35B-A3B with 40 updates of verifier-reward GRPO (Group Relative Policy Optimization).
Training
| Setting | Value |
|---|---|
| Base model | Qwen/Qwen3.6-35B-A3B |
| Training puzzles | 10,000, generated with Reasoning Gym |
| GRPO updates | 40 |
| Batch per update | 32 prompts × 8 responses = 256 responses |
| Learning rate | 1e-6 |
| Response limit | 256 tokens; thinking disabled |
| Hardware | 1 node with 8 × NVIDIA B200 GPUs |
| Training framework | Miles |
Evaluation and scope
- Development evaluation: 512 puzzles per domain. Scores are not reported in this model card.
- Held-out test set: not evaluated.
- Export: trained language-model weights. Multimodal capabilities are not validated.
- Validate independently before downstream use. Equivalence to the results reported in the Miles PR has not been established.
Each specialist was trained on its own puzzle domain. Reasoning Gym generated 10,000 training puzzles per domain.
| Component | Source revision |
|---|---|
| Base model | 995ad96eacd98c81ed38be0c5b274b04031597b0 |
Miles — radixark/miles, PR #3116 | 8f8e4dff55d80bd9d36c3490c333a119f38480ed |
| Reasoning Gym | 49b07130b3fcd12f2d064bba7c43869543a0e7e7 |
SGLang used the Triton MoE backend because automatic backend selection failed during weight transfer.
</details>Companion checkpoint: Graph Coloring
- Task
- Reinforcement Learning
- Library
- transformers
- Type
- qwen3_5_moe
- Created
- Sep 10, 2026
- Updated
- Sep 10, 2026