Instructions to use Xerv-AI/MAXWELL with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Xerv-AI/MAXWELL with Transformers:

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

pipe = pipeline("text-generation", model="Xerv-AI/MAXWELL")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Xerv-AI/MAXWELL")
model = AutoModelForCausalLM.from_pretrained("Xerv-AI/MAXWELL")
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]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Xerv-AI/MAXWELL with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/Xerv-AI/MAXWELL

SGLang

How to use Xerv-AI/MAXWELL 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 "Xerv-AI/MAXWELL" \
    --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": "Xerv-AI/MAXWELL",
		"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 "Xerv-AI/MAXWELL" \
        --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": "Xerv-AI/MAXWELL",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Studio new

How to use Xerv-AI/MAXWELL with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for Xerv-AI/MAXWELL to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for Xerv-AI/MAXWELL to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Xerv-AI/MAXWELL to start chatting

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="Xerv-AI/MAXWELL",
    max_seq_length=2048,
)

Docker Model Runner
How to use Xerv-AI/MAXWELL with Docker Model Runner:
```
docker model run hf.co/Xerv-AI/MAXWELL
```

Phase-Technologies commited on 24 days ago

Commit

58162aa

verified ·

1 Parent(s): aa390ba

Update README.md

Browse files

Files changed (1) hide show

README.md +45 -12

README.md CHANGED Viewed

@@ -33,27 +33,63 @@ model_index:
   results:
   - task:
       type: text-generation
-      name: Mathematical Reasoning
     dataset:
       name: GSM8K
       type: gsm8k
       split: test
     metrics:
     - type: accuracy
-      value: 82.4  # Data based on internal Qwen2.5-Math benchmarks
-      name: Final Synthesized Accuracy
   - task:
       type: text-generation
-      name: Physics Conceptual Logic
     dataset:
-      name: ScienceQA
-      type: science_qa
       split: test
     metrics:
     - type: accuracy
-      value: 79.1
-      name: Conceptual Rigor
 # Technical Architecture Settings
 model_type: qwen2
 quantization: 4-bit (bitsandbytes)
@@ -75,9 +111,6 @@ hardware_specification:
   vram: 16GB
   optimization: Unsloth-Fast-Inference
-# Social & Reference
-extra_gated_heading: "Phase-Technologies Proprietary Reasoning Framework"
-extra_gated_description: "Accessing this model grants permission to utilize the STSS synthesis protocols for analytical verification."
 ---

   results:
   - task:
       type: text-generation
+      name: Grade School Mathematics
     dataset:
       name: GSM8K
       type: gsm8k
       split: test
     metrics:
     - type: accuracy
+      value: 70.0
+      name: Exact Match (Zero-Shot)
   - task:
       type: text-generation
+      name: Competition Mathematics
     dataset:
+      name: MATH-Hard
+      type: lighteval/MATH-Hard
+      config: default
       split: test
     metrics:
     - type: accuracy
+      value: 60.0
+      name: Exact Match (Boxed)
+  - task:
+      type: text-generation
+      name: Professional Knowledge
+    dataset:
+      name: MMLU-Pro
+      type: TIGER-Lab/MMLU-Pro
+      config: default
+      split: test
+    metrics:
+    - type: accuracy
+      value: 45.0
+      name: Multiple Choice Accuracy
+  - task:
+      type: text-generation
+      name: Invitational Math
+    dataset:
+      name: AIME 2026
+      type: MathArena/aime_2026
+      split: train
+    metrics:
+    - type: accuracy
+      value: 10.0
+      name: Accuracy
+  - task:
+      type: text-generation
+      name: Advanced Graduate Reasoning
+    dataset:
+      name: Humanity's Last Exam
+      type: cais/hle
+      config: default
+      split: test
+    metrics:
+    - type: accuracy
+      value: 0.0
+      name: Exact String Match
 # Technical Architecture Settings
 model_type: qwen2
 quantization: 4-bit (bitsandbytes)
   vram: 16GB
   optimization: Unsloth-Fast-Inference
 ---