Instructions to use AATensorPlay/TensorMind-1.5-preview with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use AATensorPlay/TensorMind-1.5-preview with Transformers:

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

pipe = pipeline("text-generation", model="AATensorPlay/TensorMind-1.5-preview")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("AATensorPlay/TensorMind-1.5-preview")
model = AutoModelForCausalLM.from_pretrained("AATensorPlay/TensorMind-1.5-preview", 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 AATensorPlay/TensorMind-1.5-preview with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "AATensorPlay/TensorMind-1.5-preview"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "AATensorPlay/TensorMind-1.5-preview",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/AATensorPlay/TensorMind-1.5-preview

SGLang

How to use AATensorPlay/TensorMind-1.5-preview 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 "AATensorPlay/TensorMind-1.5-preview" \
    --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": "AATensorPlay/TensorMind-1.5-preview",
		"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 "AATensorPlay/TensorMind-1.5-preview" \
        --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": "AATensorPlay/TensorMind-1.5-preview",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use AATensorPlay/TensorMind-1.5-preview with Docker Model Runner:
```
docker model run hf.co/AATensorPlay/TensorMind-1.5-preview
```

TensorMind-1.5-preview / assets /benchmark-matrix.svg

TensorMind

Release TensorMind 1.5 Preview

8ce2c21 verified 8 days ago

preview code

Raw

History Blame Contribute Delete

22.4 kB

	<?xml version="1.0" encoding="utf-8" standalone="no"?>
	<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN"
	"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
	<svg xmlns:xlink="http://www.w3.org/1999/xlink" width="1008pt" height="567pt" viewBox="0 0 1008 567" xmlns="http://www.w3.org/2000/svg" version="1.1">
	<metadata>
	<rdf:RDF xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:cc="http://creativecommons.org/ns#" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#">
	<cc:Work>
	<dc:type rdf:resource="http://purl.org/dc/dcmitype/StillImage"/>
	<dc:date>2026-08-02T02:24:49.164389</dc:date>
	<dc:format>image/svg+xml</dc:format>
	<dc:creator>
	<cc:Agent>
	<dc:title>Matplotlib v3.10.9, https://matplotlib.org/</dc:title>
	</cc:Agent>
	</dc:creator>
	</cc:Work>
	</rdf:RDF>
	</metadata>
	<defs>
	<style type="text/css">*{stroke-linejoin: round; stroke-linecap: butt}</style>
	</defs>
	<g id="figure_1">
	<g id="patch_1">
	<path d="M 0 567
	L 1008 567
	L 1008 0
	L 0 0
	z
	" style="fill: #07111f"/>
	</g>
	<g id="axes_1">
	<g id="patch_2">
	<path d="M 80.64 307.875701
	L 345.24 307.875701
	L 345.24 170.1
	L 80.64 170.1
	z
	" style="fill: #0c1a2c"/>
	</g>
	<g id="patch_3">
	<path d="M 80.64 307.875701
	L 85.4028 307.875701
	L 85.4028 170.1
	L 80.64 170.1
	z
	" clip-path="url(#pc48552ebb2)" style="fill: #267bff"/>
	</g>
	<g id="matplotlib.axis_1"/>
	<g id="matplotlib.axis_2"/>
	<g id="text_1">
	<text style="font-weight: 700; font-size: 9px; font-family: 'DejaVu Sans'; text-anchor: start; fill: #8ea5c3" x="104.454" y="211.43271" transform="rotate(-0 104.454 211.43271)">CMMLU</text>
	</g>
	<g id="text_2">
	<text style="font-weight: 700; font-size: 27px; font-family: 'DejaVu Sans'; text-anchor: start; fill: #267bff" x="104.454" y="267.920748" transform="rotate(-0 104.454 267.920748)">24.8834</text>
	</g>
	<g id="text_3">
	<text style="font-size: 8.5px; font-family: 'DejaVu Sans'; text-anchor: start; fill: #8ea5c3" x="104.454" y="291.342617" transform="rotate(-0 104.454 291.342617)">accuracy</text>
	</g>
	</g>
	<g id="axes_2">
	<g id="patch_4">
	<path d="M 371.7 307.875701
	L 636.3 307.875701
	L 636.3 170.1
	L 371.7 170.1
	z
	" style="fill: #0c1a2c"/>
	</g>
	<g id="patch_5">
	<path d="M 371.7 307.875701
	L 376.4628 307.875701
	L 376.4628 170.1
	L 371.7 170.1
	z
	" clip-path="url(#p497ce62b42)" style="fill: #2ed7ff"/>
	</g>
	<g id="matplotlib.axis_3"/>
	<g id="matplotlib.axis_4"/>
	<g id="text_4">
	<text style="font-weight: 700; font-size: 9px; font-family: 'DejaVu Sans'; text-anchor: start; fill: #8ea5c3" x="395.514" y="211.43271" transform="rotate(-0 395.514 211.43271)">AGIEVAL-CN</text>
	</g>
	<g id="text_5">
	<text style="font-weight: 700; font-size: 27px; font-family: 'DejaVu Sans'; text-anchor: start; fill: #2ed7ff" x="395.514" y="267.920748" transform="rotate(-0 395.514 267.920748)">32.3822</text>
	</g>
	<g id="text_6">
	<text style="font-size: 8.5px; font-family: 'DejaVu Sans'; text-anchor: start; fill: #8ea5c3" x="395.514" y="291.342617" transform="rotate(-0 395.514 291.342617)">accuracy</text>
	</g>
	</g>
	<g id="axes_3">
	<g id="patch_6">
	<path d="M 662.76 307.875701
	L 927.36 307.875701
	L 927.36 170.1
	L 662.76 170.1
	z
	" style="fill: #0c1a2c"/>
	</g>
	<g id="patch_7">
	<path d="M 662.76 307.875701
	L 667.5228 307.875701
	L 667.5228 170.1
	L 662.76 170.1
	z
	" clip-path="url(#p0bbd4284d2)" style="fill: #267bff"/>
	</g>
	<g id="matplotlib.axis_5"/>
	<g id="matplotlib.axis_6"/>
	<g id="text_7">
	<text style="font-weight: 700; font-size: 9px; font-family: 'DejaVu Sans'; text-anchor: start; fill: #8ea5c3" x="686.574" y="211.43271" transform="rotate(-0 686.574 211.43271)">A-CLUE</text>
	</g>
	<g id="text_8">
	<text style="font-weight: 700; font-size: 27px; font-family: 'DejaVu Sans'; text-anchor: start; fill: #267bff" x="686.574" y="267.920748" transform="rotate(-0 686.574 267.920748)">24.7282</text>
	</g>
	<g id="text_9">
	<text style="font-size: 8.5px; font-family: 'DejaVu Sans'; text-anchor: start; fill: #8ea5c3" x="686.574" y="291.342617" transform="rotate(-0 686.574 291.342617)">accuracy</text>
	</g>
	</g>
	<g id="axes_4">
	<g id="patch_8">
	<path d="M 80.64 464.94
	L 345.24 464.94
	L 345.24 327.164299
	L 80.64 327.164299
	z
	" style="fill: #0c1a2c"/>
	</g>
	<g id="patch_9">
	<path d="M 80.64 464.94
	L 85.4028 464.94
	L 85.4028 327.164299
	L 80.64 327.164299
	z
	" clip-path="url(#p9de0fb1870)" style="fill: #ff9d42"/>
	</g>
	<g id="matplotlib.axis_7"/>
	<g id="matplotlib.axis_8"/>
	<g id="text_10">
	<text style="font-weight: 700; font-size: 9px; font-family: 'DejaVu Sans'; text-anchor: start; fill: #8ea5c3" x="104.454" y="368.497009" transform="rotate(-0 104.454 368.497009)">C-EVAL</text>
	</g>
	<g id="text_11">
	<text style="font-weight: 700; font-size: 27px; font-family: 'DejaVu Sans'; text-anchor: start; fill: #ff9d42" x="104.454" y="424.985047" transform="rotate(-0 104.454 424.985047)">23.2541</text>
	</g>
	<g id="text_12">
	<text style="font-size: 8.5px; font-family: 'DejaVu Sans'; text-anchor: start; fill: #8ea5c3" x="104.454" y="448.406916" transform="rotate(-0 104.454 448.406916)">accuracy</text>
	</g>
	</g>
	<g id="axes_5">
	<g id="patch_10">
	<path d="M 371.7 464.94
	L 636.3 464.94
	L 636.3 327.164299
	L 371.7 327.164299
	z
	" style="fill: #0c1a2c"/>
	</g>
	<g id="patch_11">
	<path d="M 371.7 464.94
	L 376.4628 464.94
	L 376.4628 327.164299
	L 371.7 327.164299
	z
	" clip-path="url(#p635785d11b)" style="fill: #2ed7ff"/>
	</g>
	<g id="matplotlib.axis_9"/>
	<g id="matplotlib.axis_10"/>
	<g id="text_13">
	<text style="font-weight: 700; font-size: 9px; font-family: 'DejaVu Sans'; text-anchor: start; fill: #8ea5c3" x="395.514" y="368.497009" transform="rotate(-0 395.514 368.497009)">TMMLU+</text>
	</g>
	<g id="text_14">
	<text style="font-weight: 700; font-size: 27px; font-family: 'DejaVu Sans'; text-anchor: start; fill: #2ed7ff" x="395.514" y="424.985047" transform="rotate(-0 395.514 424.985047)">24.7272</text>
	</g>
	<g id="text_15">
	<text style="font-size: 8.5px; font-family: 'DejaVu Sans'; text-anchor: start; fill: #8ea5c3" x="395.514" y="448.406916" transform="rotate(-0 395.514 448.406916)">accuracy</text>
	</g>
	</g>
	<g id="axes_6">
	<g id="patch_12">
	<path d="M 662.76 464.94
	L 927.36 464.94
	L 927.36 327.164299
	L 662.76 327.164299
	z
	" style="fill: #10243a"/>
	</g>
	<g id="patch_13">
	<path d="M 662.76 464.94
	L 667.5228 464.94
	L 667.5228 327.164299
	L 662.76 327.164299
	z
	" clip-path="url(#p13a9443a99)" style="fill: #f4f8ff"/>
	</g>
	<g id="matplotlib.axis_11"/>
	<g id="matplotlib.axis_12"/>
	<g id="text_16">
	<text style="font-weight: 700; font-size: 9px; font-family: 'DejaVu Sans'; text-anchor: start; fill: #8ea5c3" x="686.574" y="368.497009" transform="rotate(-0 686.574 368.497009)">5-SUITE MACRO</text>
	</g>
	<g id="text_17">
	<text style="font-weight: 700; font-size: 27px; font-family: 'DejaVu Sans'; text-anchor: start; fill: #f4f8ff" x="686.574" y="424.985047" transform="rotate(-0 686.574 424.985047)">25.9950</text>
	</g>
	<g id="text_18">
	<text style="font-size: 8.5px; font-family: 'DejaVu Sans'; text-anchor: start; fill: #8ea5c3" x="686.574" y="448.406916" transform="rotate(-0 686.574 448.406916)">accuracy</text>
	</g>
	</g>
	<g id="AnnotationBbox_1">
	<image xlink:href="data:image/png;base64,
	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" id="image3d08965066" transform="scale(1 -1) translate(0 -95.4)" x="792.96" y="0.249" width="107.64" height="95.4"/>
	</g>
	<g id="text_19">
	<text style="font-weight: 700; font-size: 13px; font-family: 'DejaVu Sans'; text-anchor: start; fill: #2ed7ff" x="55.44" y="53.865" transform="rotate(-0 55.44 53.865)">TensorMind 1.5 Preview</text>
	</g>
	<g id="text_20">
	<text style="font-weight: 700; font-size: 28px; font-family: 'DejaVu Sans'; text-anchor: start; fill: #f4f8ff" x="55.44" y="89.019" transform="rotate(-0 55.44 89.019)">Benchmark scorecard</text>
	</g>
	<g id="text_21">
	<text style="font-size: 11px; font-family: 'DejaVu Sans'; text-anchor: start; fill: #8ea5c3" x="55.44" y="117.936" transform="rotate(-0 55.44 117.936)">TensorMind 1.5 Preview · full-dataset accuracy (%)</text>
	</g>
	<g id="text_22">
	<text style="font-size: 8.5px; font-family: 'DejaVu Sans'; text-anchor: start; fill: #8ea5c3" x="55.44" y="535.815" transform="rotate(-0 55.44 535.815)">Protocol lm-eval 0.4.12 · SGLang 0.5.14 · 0-shot · full datasets · batch 48 · fixed seeds</text>
	</g>
	<g id="text_23">
	<text style="font-size: 8.5px; font-family: 'DejaVu Sans'; text-anchor: end; fill: #8ea5c3" x="962.64" y="535.815" transform="rotate(-0 962.64 535.815)">Accuracy, higher is better</text>
	</g>
	</g>
	<defs>
	<clipPath id="pc48552ebb2">
	<rect x="80.64" y="170.1" width="264.6" height="137.775701"/>
	</clipPath>
	<clipPath id="p497ce62b42">
	<rect x="371.7" y="170.1" width="264.6" height="137.775701"/>
	</clipPath>
	<clipPath id="p0bbd4284d2">
	<rect x="662.76" y="170.1" width="264.6" height="137.775701"/>
	</clipPath>
	<clipPath id="p9de0fb1870">
	<rect x="80.64" y="327.164299" width="264.6" height="137.775701"/>
	</clipPath>
	<clipPath id="p635785d11b">
	<rect x="371.7" y="327.164299" width="264.6" height="137.775701"/>
	</clipPath>
	<clipPath id="p13a9443a99">
	<rect x="662.76" y="327.164299" width="264.6" height="137.775701"/>
	</clipPath>
	</defs>
	</svg>