Instructions to use Premchan369/Q-TensorFormer with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Premchan369/Q-TensorFormer with Transformers:

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

pipe = pipeline("text-generation", model="Premchan369/Q-TensorFormer", trust_remote_code=True)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("Premchan369/Q-TensorFormer", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Premchan369/Q-TensorFormer with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Premchan369/Q-TensorFormer"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Premchan369/Q-TensorFormer",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Premchan369/Q-TensorFormer

SGLang

How to use Premchan369/Q-TensorFormer 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 "Premchan369/Q-TensorFormer" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Premchan369/Q-TensorFormer",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "Premchan369/Q-TensorFormer" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Premchan369/Q-TensorFormer",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Premchan369/Q-TensorFormer with Docker Model Runner:
```
docker model run hf.co/Premchan369/Q-TensorFormer
```

Q-TensorFormer / outputs /controller_convergence.json

Premchandyadav369

feat(research): Turn Q-TensorFormer into a verified research system with LFS figure assets

0e4850b 26 days ago

Raw History Blame Contribute Delete

10.5 kB

	{
	"metadata": {
	"timestamp": "2026-09-11 17:39:39",
	"total_steps": 100,
	"step_disturbance_step": 30,
	"target_latency_ms": 1.2
	},
	"controllers": {
	"PID_Dual_Controller": {
	"settling_steps_post_disturbance": 70,
	"violation_rate_pct": 64.29,
	"max_overshoot_pct": 35.67,
	"steady_state_error_ms": 0.4932,
	"mean_rank_post_disturbance": 2.93,
	"history_sample": [
	{
	"step": 1,
	"seq_len": 32,
	"chosen_rank": 4,
	"measured_latency_ms": 0.91,
	"target_latency_ms": 1.2,
	"error": -0.29,
	"lambda_latency": 0.9681
	},
	{
	"step": 11,
	"seq_len": 32,
	"chosen_rank": 4,
	"measured_latency_ms": 0.891,
	"target_latency_ms": 1.2,
	"error": -0.309,
	"lambda_latency": 0.355
	},
	{
	"step": 21,
	"seq_len": 32,
	"chosen_rank": 4,
	"measured_latency_ms": 0.929,
	"target_latency_ms": 1.2,
	"error": -0.271,
	"lambda_latency": 0.05
	},
	{
	"step": 31,
	"seq_len": 128,
	"chosen_rank": 4,
	"measured_latency_ms": 1.579,
	"target_latency_ms": 1.2,
	"error": 0.379,
	"lambda_latency": 0.05
	},
	{
	"step": 41,
	"seq_len": 128,
	"chosen_rank": 4,
	"measured_latency_ms": 1.606,
	"target_latency_ms": 1.2,
	"error": 0.406,
	"lambda_latency": 0.0766
	},
	{
	"step": 51,
	"seq_len": 128,
	"chosen_rank": 4,
	"measured_latency_ms": 1.598,
	"target_latency_ms": 1.2,
	"error": 0.398,
	"lambda_latency": 0.6596
	},
	{
	"step": 61,
	"seq_len": 128,
	"chosen_rank": 4,
	"measured_latency_ms": 1.582,
	"target_latency_ms": 1.2,
	"error": 0.382,
	"lambda_latency": 1.9008
	},
	{
	"step": 71,
	"seq_len": 128,
	"chosen_rank": 4,
	"measured_latency_ms": 1.598,
	"target_latency_ms": 1.2,
	"error": 0.398,
	"lambda_latency": 3.215
	},
	{
	"step": 81,
	"seq_len": 128,
	"chosen_rank": 1,
	"measured_latency_ms": 0.703,
	"target_latency_ms": 1.2,
	"error": -0.497,
	"lambda_latency": 3.8861
	},
	{
	"step": 91,
	"seq_len": 128,
	"chosen_rank": 1,
	"measured_latency_ms": 0.709,
	"target_latency_ms": 1.2,
	"error": -0.491,
	"lambda_latency": 3.3584
	}
	]
	},
	"P_Only_Controller": {
	"settling_steps_post_disturbance": 70,
	"violation_rate_pct": 82.86,
	"max_overshoot_pct": 35.75,
	"steady_state_error_ms": 0.4175,
	"mean_rank_post_disturbance": 3.49,
	"history_sample": [
	{
	"step": 1,
	"seq_len": 32,
	"chosen_rank": 1,
	"measured_latency_ms": 0.372,
	"target_latency_ms": 1.2,
	"error": -0.828,
	"lambda_latency": 0.9337
	},
	{
	"step": 11,
	"seq_len": 32,
	"chosen_rank": 4,
	"measured_latency_ms": 0.862,
	"target_latency_ms": 1.2,
	"error": -0.338,
	"lambda_latency": 0.6921
	},
	{
	"step": 21,
	"seq_len": 32,
	"chosen_rank": 4,
	"measured_latency_ms": 0.916,
	"target_latency_ms": 1.2,
	"error": -0.284,
	"lambda_latency": 0.4587
	},
	{
	"step": 31,
	"seq_len": 128,
	"chosen_rank": 4,
	"measured_latency_ms": 1.56,
	"target_latency_ms": 1.2,
	"error": 0.36,
	"lambda_latency": 0.2713
	},
	{
	"step": 41,
	"seq_len": 128,
	"chosen_rank": 4,
	"measured_latency_ms": 1.596,
	"target_latency_ms": 1.2,
	"error": 0.396,
	"lambda_latency": 0.5824
	},
	{
	"step": 51,
	"seq_len": 128,
	"chosen_rank": 4,
	"measured_latency_ms": 1.596,
	"target_latency_ms": 1.2,
	"error": 0.396,
	"lambda_latency": 0.8945
	},
	{
	"step": 61,
	"seq_len": 128,
	"chosen_rank": 4,
	"measured_latency_ms": 1.572,
	"target_latency_ms": 1.2,
	"error": 0.372,
	"lambda_latency": 1.2079
	},
	{
	"step": 71,
	"seq_len": 128,
	"chosen_rank": 1,
	"measured_latency_ms": 0.689,
	"target_latency_ms": 1.2,
	"error": -0.511,
	"lambda_latency": 1.3141
	},
	{
	"step": 81,
	"seq_len": 128,
	"chosen_rank": 4,
	"measured_latency_ms": 1.604,
	"target_latency_ms": 1.2,
	"error": 0.404,
	"lambda_latency": 1.423
	},
	{
	"step": 91,
	"seq_len": 128,
	"chosen_rank": 1,
	"measured_latency_ms": 0.698,
	"target_latency_ms": 1.2,
	"error": -0.502,
	"lambda_latency": 1.4477
	}
	]
	},
	"Fixed_Multiplier": {
	"settling_steps_post_disturbance": 70,
	"violation_rate_pct": 100.0,
	"max_overshoot_pct": 36.17,
	"steady_state_error_ms": 0.3943,
	"mean_rank_post_disturbance": 4.0,
	"history_sample": [
	{
	"step": 1,
	"seq_len": 32,
	"chosen_rank": 4,
	"measured_latency_ms": 0.907,
	"target_latency_ms": 1.2,
	"error": -0.293,
	"lambda_latency": 1.0
	},
	{
	"step": 11,
	"seq_len": 32,
	"chosen_rank": 4,
	"measured_latency_ms": 0.911,
	"target_latency_ms": 1.2,
	"error": -0.289,
	"lambda_latency": 1.0
	},
	{
	"step": 21,
	"seq_len": 32,
	"chosen_rank": 4,
	"measured_latency_ms": 0.946,
	"target_latency_ms": 1.2,
	"error": -0.254,
	"lambda_latency": 1.0
	},
	{
	"step": 31,
	"seq_len": 128,
	"chosen_rank": 4,
	"measured_latency_ms": 1.577,
	"target_latency_ms": 1.2,
	"error": 0.377,
	"lambda_latency": 1.0
	},
	{
	"step": 41,
	"seq_len": 128,
	"chosen_rank": 4,
	"measured_latency_ms": 1.575,
	"target_latency_ms": 1.2,
	"error": 0.375,
	"lambda_latency": 1.0
	},
	{
	"step": 51,
	"seq_len": 128,
	"chosen_rank": 4,
	"measured_latency_ms": 1.566,
	"target_latency_ms": 1.2,
	"error": 0.366,
	"lambda_latency": 1.0
	},
	{
	"step": 61,
	"seq_len": 128,
	"chosen_rank": 4,
	"measured_latency_ms": 1.573,
	"target_latency_ms": 1.2,
	"error": 0.373,
	"lambda_latency": 1.0
	},
	{
	"step": 71,
	"seq_len": 128,
	"chosen_rank": 4,
	"measured_latency_ms": 1.62,
	"target_latency_ms": 1.2,
	"error": 0.42,
	"lambda_latency": 1.0
	},
	{
	"step": 81,
	"seq_len": 128,
	"chosen_rank": 4,
	"measured_latency_ms": 1.593,
	"target_latency_ms": 1.2,
	"error": 0.393,
	"lambda_latency": 1.0
	},
	{
	"step": 91,
	"seq_len": 128,
	"chosen_rank": 4,
	"measured_latency_ms": 1.587,
	"target_latency_ms": 1.2,
	"error": 0.387,
	"lambda_latency": 1.0
	}
	]
	},
	"Static_Heuristic": {
	"settling_steps_post_disturbance": 70,
	"violation_rate_pct": 100.0,
	"max_overshoot_pct": 124.67,
	"steady_state_error_ms": 1.4516,
	"mean_rank_post_disturbance": 8.0,
	"history_sample": [
	{
	"step": 1,
	"seq_len": 32,
	"chosen_rank": 8,
	"measured_latency_ms": 1.483,
	"target_latency_ms": 1.2,
	"error": 0.283,
	"lambda_latency": 1.0
	},
	{
	"step": 11,
	"seq_len": 32,
	"chosen_rank": 8,
	"measured_latency_ms": 1.496,
	"target_latency_ms": 1.2,
	"error": 0.296,
	"lambda_latency": 1.0
	},
	{
	"step": 21,
	"seq_len": 32,
	"chosen_rank": 8,
	"measured_latency_ms": 1.502,
	"target_latency_ms": 1.2,
	"error": 0.302,
	"lambda_latency": 1.0
	},
	{
	"step": 31,
	"seq_len": 128,
	"chosen_rank": 8,
	"measured_latency_ms": 2.652,
	"target_latency_ms": 1.2,
	"error": 1.452,
	"lambda_latency": 1.0
	},
	{
	"step": 41,
	"seq_len": 128,
	"chosen_rank": 8,
	"measured_latency_ms": 2.636,
	"target_latency_ms": 1.2,
	"error": 1.436,
	"lambda_latency": 1.0
	},
	{
	"step": 51,
	"seq_len": 128,
	"chosen_rank": 8,
	"measured_latency_ms": 2.658,
	"target_latency_ms": 1.2,
	"error": 1.458,
	"lambda_latency": 1.0
	},
	{
	"step": 61,
	"seq_len": 128,
	"chosen_rank": 8,
	"measured_latency_ms": 2.662,
	"target_latency_ms": 1.2,
	"error": 1.462,
	"lambda_latency": 1.0
	},
	{
	"step": 71,
	"seq_len": 128,
	"chosen_rank": 8,
	"measured_latency_ms": 2.652,
	"target_latency_ms": 1.2,
	"error": 1.452,
	"lambda_latency": 1.0
	},
	{
	"step": 81,
	"seq_len": 128,
	"chosen_rank": 8,
	"measured_latency_ms": 2.635,
	"target_latency_ms": 1.2,
	"error": 1.435,
	"lambda_latency": 1.0
	},
	{
	"step": 91,
	"seq_len": 128,
	"chosen_rank": 8,
	"measured_latency_ms": 2.634,
	"target_latency_ms": 1.2,
	"error": 1.434,
	"lambda_latency": 1.0
	}
	]
	}
	}
	}