Instructions to use ABDUL-HASEEB-TANOLI/HAIDER-Math-32B-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ABDUL-HASEEB-TANOLI/HAIDER-Math-32B-v1 with Transformers:

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

pipe = pipeline("text-generation", model="ABDUL-HASEEB-TANOLI/HAIDER-Math-32B-v1")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ABDUL-HASEEB-TANOLI/HAIDER-Math-32B-v1")
model = AutoModelForCausalLM.from_pretrained("ABDUL-HASEEB-TANOLI/HAIDER-Math-32B-v1", 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 ABDUL-HASEEB-TANOLI/HAIDER-Math-32B-v1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ABDUL-HASEEB-TANOLI/HAIDER-Math-32B-v1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ABDUL-HASEEB-TANOLI/HAIDER-Math-32B-v1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ABDUL-HASEEB-TANOLI/HAIDER-Math-32B-v1

SGLang

How to use ABDUL-HASEEB-TANOLI/HAIDER-Math-32B-v1 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 "ABDUL-HASEEB-TANOLI/HAIDER-Math-32B-v1" \
    --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": "ABDUL-HASEEB-TANOLI/HAIDER-Math-32B-v1",
		"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 "ABDUL-HASEEB-TANOLI/HAIDER-Math-32B-v1" \
        --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": "ABDUL-HASEEB-TANOLI/HAIDER-Math-32B-v1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use ABDUL-HASEEB-TANOLI/HAIDER-Math-32B-v1 with Docker Model Runner:
```
docker model run hf.co/ABDUL-HASEEB-TANOLI/HAIDER-Math-32B-v1
```

"Can HAIDER-Math-32B Beat Top Math Models? Community Test & Feedback Thread”

by ABDUL-HASEEB-TANOLI - opened May 1

Discussion

ABDUL-HASEEB-TANOLI

Owner May 1

Hi everyone,

I’ve recently released HAIDER-Math-32B (Original), a model specifically designed for advanced mathematical reasoning. Early traction has been encouraging, and I’d like to invite the community to rigorously evaluate its performance.

🔍 Focus of evaluation:

Mathematical reasoning (not general chat)
Step-by-step solution quality
Final answer accuracy
Logical consistency across multi-step problems

🧪 Recommended test areas:

GSM8K-style word problems
Algebra and equation solving
Multi-step arithmetic
Edge-case and tricky reasoning problems

📊 If you test the model, please share:

The questions used
Model outputs
Correct answers
(Optional) Comparison with other strong math models

⚠️ Notes:

This is the original (full-precision) model
Results from this version will be considered the primary benchmark reference

🎯 Goal:
To evaluate whether this model can match or surpass strong mathematical reasoning baselines through open and transparent community testing.

I will compile and summarize community findings into a structured benchmark report.

Thanks in advance — your feedback will directly contribute to improving and validating the model.

ABDUL-HASEEB-TANOLI changed discussion status to closed May 1

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment