Downloads · 30 days
17
16% of all-time downloads
ianlee1996/pokerbench-qwen3-14b-lora-mixed-v2
pokerbench-qwen3-14b-lora-mixed-v2 is a text generation model from ianlee1996. Use it when you need the model to write or continue text. It is set up for peft. The card lists the license as apache-2.0.
Successor to pokerbench-qwen3-14b-lora-mixed. Same training recipe (50/50 paper-format + production-PE-format, 250 steps on top of the Stage 1 paper LoRA), with one critical bug fix:
Downloads · 30 days
17
16% of all-time downloads
All-time downloads
104
Public
Repo size
525 MB
Likes
1
Public
Click a slice to open those files.
.safetensors514 MB · 98%
From the Hugging Face model README
Successor to pokerbench-qwen3-14b-lora-mixed. Same training recipe (50/50 paper-format + production-PE-format, 250 steps on top of the Stage 1 paper LoRA), with one critical bug fix:
Fixed: PE training labels for
raise/betactions previously dropped the sizing — every raise label was the bare string"raise". The model trained on this never learned to emit a raise amount, so production deployments sawdefault_pct: Nonewhenever the model chose to raise. The fixed PE jsonl recovers the BB amount from the prompt'sraise pct=N(Mbb,...)line and rewrites the label as"raise N".
Old mixed (v1) | This (v2 sized) | |
|---|---|---|
| Raise output | "raise" (1 token, no sizing) | "raise 14.0" (proper sizing) |
| Strict raise sizing on PE 60-sample eval | 0/60 (0.0%) | 41/60 (68.3%) |
| Paper 11k EM | 89.86% | (within ±2 expected) |
| PE 200 AA | 84.0% | 84.0% |
The action-accuracy headline is unchanged because the eval scoring folds raise/bet/all-in into a single "aggressive" kind. The sizing is what production deployment cares about.
gen-pe-format-dataset.ts used row.correctDecision (action kind only) as the label. The fix is post-hoc — scripts/fix_pe_raise_sizing.py reads each prompt, extracts BB amount from the raise pct=N(Mbb, ~M) / raise pct=N(Bet M, ~M) line, and rewrites the label.
After fix:
"raise" remainingraise 14.0, raise 13.0, raise 25.0, raise 2.5, raise 10.0)prompt: "...preflop SB facing UTG raise 2.5..."
v1 (old): "raise" ← parser sees no number, default_pct: None
v2 (this): "raise 14.5" ← parser gets 14.5 BB, action complete
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
base = AutoModelForCausalLM.from_pretrained(
"Qwen/Qwen3-14B",
torch_dtype=torch.bfloat16,
device_map="auto",
)
model = PeftModel.from_pretrained(base, "ianlee1996/pokerbench-qwen3-14b-lora-mixed-v2")
tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen3-14B")
system_prompt = (
"You are a specialist in playing 6-handed No Limit Texas Holdem. "
"Output ONLY the optimal action with no explanation. "
"Valid formats: 'fold', 'check', 'call', 'bet N', 'raise N', 'all-in'."
)
# Inference at temp 0.1, top-p 0.95, max_tokens=16 (raise output is 5-6 tokens)
Qwen/Qwen3-14B04_qwen3_14b_lora_full_data_1epoch/checkpoint-4375 (60k+500k paper-format SFT, 1 epoch, EM 90.07%)Full pipeline at https://github.com/IanLiYi1996/PokerBench:
# 1. Fix the existing PE jsonl in-place
.venv/bin/python -m scripts.fix_pe_raise_sizing \
data/pe_format/mixed_paper_pe_50_50.jsonl \
data/pe_format/mixed_paper_pe_50_50_fixed.jsonl
# 2. Re-train the SFT mixed checkpoint on the fixed data
.venv/bin/python -m scripts.train --config configs/experiments/09_qwen3_14b_lora_mixed_v2_sized.yaml
@inproceedings{zhuang2025pokerbench,
title={PokerBench: Training Large Language Models to become Professional Poker Players},
author={Zhuang, Richard and Gupta, Akshat and Yang, Richard and Rahane, Aniket and Li, Zhengyu and Anumanchipalli, Gopala},
booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},
year={2025},
url={https://arxiv.org/abs/2501.08328}
}
Apache-2.0, matching Qwen/Qwen3-14B and the PokerBench dataset.