Downloads · 30 days
7
9% of all-time downloads
qwen3-8b-openspiel-mix8-curriculum-100iter
GoodStartLabs/qwen3-8b-openspiel-mix8-curriculum-100iter
qwen3-8b-openspiel-mix8-curriculum-100iter is a reinforcement learning model from GoodStartLabs. 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 peft.
LoRA adapter for Qwen/Qwen3-8B trained with Tinker / tinker-cookbook on an 8-game OpenSpiel mix using a 3-phase scripted-opponent curriculum.
All-time downloads
74
Public
Repo size
738 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors369 MB · 100%
At a glance
- Task
- Reinforcement Learning
- Library
- peft
- Access
- Public
- Created
- Apr 27, 2026
- Updated
- Apr 29, 2026
- SHA
- 2d8215dd
Base models
Model card
From the Hugging Face model README
qwen3-8b-openspiel-mix8-curriculum-100iter
LoRA adapter for Qwen/Qwen3-8B trained with Tinker /
tinker-cookbook on an 8-game
OpenSpiel mix using a 3-phase scripted-opponent curriculum.
Run
| Run name | openspiel-qwen3-8b-mix8-curriculum-100iter-202604260230 |
| Wandb | good-start-labs/Neurips 2026 — run name openspiel-qwen3-8b-mix8-curriculum-100iter-202604260230 |
| Tinker train id (final) | 884edcc2-4da0-5569-a16f-8f1fec64914e |
| Source checkpoint | tinker://884edcc2-4da0-5569-a16f-8f1fec64914e:train:0/sampler_weights/final |
| Date | 2026-04-26 |
Note:
checkpoints.jsonlfor this run contains twofinalentries due to a supervisor restart at iter 100; this adapter is from the last entry (884edcc2-4da0-5569-a16f-8f1fec64914e), the run that completed cleanly.
Training config
- Base model:
Qwen/Qwen3-8B - Renderer:
qwen3_disable_thinking(prefills<think></think>) - LoRA: rank 32, alpha 32, targets
{q,k,v,o,gate,up,down}_proj+lm_head - Optim: lr
1e-5, importance-sampling loss, KL coef 0 - Rollouts:
groups_per_batch=8,group_size=4,max_tokens=32,temperature=1.0 - Schedule:
n_batches=100,eval_every=10,save_every=10(checkpoints 10/20/.../100/final) - Execution: stream, 2 minibatches, 1 substep
- Returns:
normalize_returns=True,invalid_action_penalty=-1.0 - Self-play: disabled
Game mix and opponent curricula
mixture_mode="round_robin", seat_mode="uniform", transform_simultaneous_to_turn_based=True.
Multi-player games use a 3-phase curriculum keyed on training_fraction = batch_idx / (n_batches - 1),
with phase boundaries at 1/3 and 2/3.
| Game | Players | Curriculum (easy / medium / hard) |
|---|---|---|
blackjack | 1p + chance | — |
morpion_solitaire | 1p combinatorial | — |
nim (misere, piles 1/3/5/7) | 2p | random / nim_greedy / nim_epsilon_0.1 |
hanabi | 2p coop, partial info | random / hanabi_rule_bot / hanabi_rule_bot |
sheriff | 2p bargaining | random / sheriff_simple / sheriff_simple |
connect_four | 2p zero-sum | random / one_step_win / win_or_block |
breakthrough | 2p zero-sum | random / one_step_win / win_or_block |
pentago | 2p zero-sum | random / one_step_win / one_step_win |
System prompt constrains output to ACTION: <index>.
Relation to sibling run
GoodStartLabs/qwen3-8b-openspiel-mix8-curriculum3-100iter uses the same config with
group_size=8 instead of 4 (i.e. 2x rollouts per batch). This run is the lower-rollout variant.
Loading
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer
base = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-8B", torch_dtype="bfloat16", device_map="auto")
tok = AutoTokenizer.from_pretrained("Qwen/Qwen3-8B")
model = PeftModel.from_pretrained(base, "GoodStartLabs/qwen3-8b-openspiel-mix8-curriculum-100iter")
- Task
- Reinforcement Learning
- Library
- peft
- Created
- Apr 27, 2026
- Updated
- Apr 29, 2026