Instructions to use promotion/PROSPER-Qwen2.5-7B-WildChecklists-2ep with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use promotion/PROSPER-Qwen2.5-7B-WildChecklists-2ep with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="promotion/PROSPER-Qwen2.5-7B-WildChecklists-2ep")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("promotion/PROSPER-Qwen2.5-7B-WildChecklists-2ep")
model = AutoModelForCausalLM.from_pretrained("promotion/PROSPER-Qwen2.5-7B-WildChecklists-2ep", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use promotion/PROSPER-Qwen2.5-7B-WildChecklists-2ep with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "promotion/PROSPER-Qwen2.5-7B-WildChecklists-2ep"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "promotion/PROSPER-Qwen2.5-7B-WildChecklists-2ep",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/promotion/PROSPER-Qwen2.5-7B-WildChecklists-2ep

SGLang

How to use promotion/PROSPER-Qwen2.5-7B-WildChecklists-2ep with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "promotion/PROSPER-Qwen2.5-7B-WildChecklists-2ep" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "promotion/PROSPER-Qwen2.5-7B-WildChecklists-2ep",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "promotion/PROSPER-Qwen2.5-7B-WildChecklists-2ep" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "promotion/PROSPER-Qwen2.5-7B-WildChecklists-2ep",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use promotion/PROSPER-Qwen2.5-7B-WildChecklists-2ep with Docker Model Runner:
```
docker model run hf.co/promotion/PROSPER-Qwen2.5-7B-WildChecklists-2ep
```

PROSPER (max-min Blackwell approachability)

Research checkpoint. Absolute max-min aggregation over the same objective-wise game values, per prompt: our reimplementation of the comparison method.

Fine-tuned from Qwen/Qwen2.5-7B-Instruct. This repository is one of three arms trained and evaluated under one protocol so that the only difference between them is the aggregation rule; the other two are linked below.

Training setting

Retrained under the protocol of "Back to Blackwell" (PROSPER) on WildChecklists, where every prompt carries its own checklist items and the judge scores one item at a time.


Prompts	522 (union-filtered so all three arms train on one prompt set)
Learner pairs	14,616 (all 28 unordered pairs of 8 sampled responses per prompt)
Pipeline epochs	2
Optimizer steps	228 at batch size 128
Learning rate	3e-7, weight decay 1e-6, AdamW, warmup ratio 0.1, grad clip 1.0
Max sequence / prompt length	2048 / 1024
Seed	555134
Candidate decoding	temperature 0.8, top_p 0.9, 8 responses per prompt
Judge for training signal	local open-weight judge, 10 judgments per pair averaged over both presentation orders

Provenance recorded by the run: dataset manifest b84df1cb1e514499, solver artifact not recorded.

Evaluation

Greedy generation (temperature 0) against each benchmark's released baseline answers, judged by a local open-weight judge at temperature 0 in both presentation orders with ties counted as 0.5.

Arm	Arena-Hard	AlpacaEval
NBPO	0.6333	0.4124
PROSPER	0.6225	0.4071
Fixed-reference Nash	0.6111	0.4118

This checkpoint: Arena-Hard 0.6225 (95% prompt bootstrap [0.5936, 0.6498], 494 prompts); AlpacaEval 0.4071 (95% prompt bootstrap [0.3893, 0.4246], 802 prompts).

Two limits these numbers carry

They are not leaderboard scores. Arena-Hard normally judges with gpt-4-1106 and AlpacaEval 2.0 with weighted_alpaca_eval_gpt4_turbo. To keep paid-API cost at zero, judging here uses a local open-weight model. The questions and baseline answers are the official static files, so the three arms are comparable with each other, but the absolute values are not comparable with published leaderboard numbers. Re-judging the same responses with a larger open-weight judge moved every arm up by about 0.11 and changed the ranking, which is why only within-judge comparisons are reported.
The differences between the trained arms are not statistically resolved. Paired whole-prompt bootstraps on the common prompts give NBPO minus PROSPER +0.0087 [-0.0122, +0.0305] on Arena-Hard and +0.0063 [-0.0087, +0.0206] on AlpacaEval. NBPO minus fixed-reference Nash does exclude zero on Arena-Hard (+0.0243 [+0.0040, +0.0446]), but that arm also fell -0.0250 below its own single-epoch run, so the gap reflects the fixed reference degrading rather than NBPO improving. General-capability benchmarks (MMLU, ARC-Challenge, HellaSwag, IFEval) separate none of the arms from each other or from the base by more than their standard errors.

The three arms

The tokenizer is unchanged from the base model.

Downloads last month: 365

Safetensors

Model size

8B params

Tensor type

BF16

Model tree for promotion/PROSPER-Qwen2.5-7B-WildChecklists-2ep

Base model

Qwen/Qwen2.5-7B

Finetuned

Qwen/Qwen2.5-7B-Instruct

Finetuned

(3100)

this model