Instructions to use physicsrob/torchwright-calculator-advanced-max-digits-12 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use physicsrob/torchwright-calculator-advanced-max-digits-12 with Transformers:

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

pipe = pipeline("text-generation", model="physicsrob/torchwright-calculator-advanced-max-digits-12")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("physicsrob/torchwright-calculator-advanced-max-digits-12")
model = AutoModelForCausalLM.from_pretrained("physicsrob/torchwright-calculator-advanced-max-digits-12", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use physicsrob/torchwright-calculator-advanced-max-digits-12 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "physicsrob/torchwright-calculator-advanced-max-digits-12"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "physicsrob/torchwright-calculator-advanced-max-digits-12",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/physicsrob/torchwright-calculator-advanced-max-digits-12

SGLang

How to use physicsrob/torchwright-calculator-advanced-max-digits-12 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 "physicsrob/torchwright-calculator-advanced-max-digits-12" \
    --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": "physicsrob/torchwright-calculator-advanced-max-digits-12",
		"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 "physicsrob/torchwright-calculator-advanced-max-digits-12" \
        --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": "physicsrob/torchwright-calculator-advanced-max-digits-12",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use physicsrob/torchwright-calculator-advanced-max-digits-12 with Docker Model Runner:
```
docker model run hf.co/physicsrob/torchwright-calculator-advanced-max-digits-12
```

torchwright-calculator-advanced-max-digits-12 / torchwright_schematic_v1.schema.json

physicsrob

Upload folder using huggingface_hub

161b4bc verified 10 days ago

Raw

History Blame Contribute Delete

3.68 kB

	{
	"$schema": "https://json-schema.org/draft/2020-12/schema",
	"$id": "https://github.com/physicsrob/torchwright/blob/main/torchwright/compiler/torchwright_schematic_v1.schema.json",
	"title": "torchwright compiled artifact schematic v1",
	"type": "object",
	"required": [
	"format",
	"integrity",
	"artifact",
	"build",
	"model",
	"token_io",
	"graphs",
	"schedule",
	"residual_stream",
	"physical_layout",
	"parameter_support",
	"observability",
	"runtime_contract",
	"intervention_contract",
	"reference_fixtures"
	],
	"properties": {
	"format": {"const": "torchwright.schematic.v1"},
	"$schema": {"type": "string"},
	"integrity": {
	"type": "object",
	"required": ["sha256", "scope"],
	"properties": {
	"sha256": {"type": "string", "pattern": "^[0-9a-f]{64}$"},
	"scope": {"type": "string"}
	}
	},
	"artifact": {
	"type": "object",
	"required": ["kind", "architecture", "files"],
	"properties": {
	"kind": {"const": "huggingface_causal_lm_bundle"},
	"architecture": {"enum": ["phi3", "custom"]},
	"files": {
	"type": "object",
	"additionalProperties": {
	"type": "object",
	"required": ["sha256", "bytes"],
	"properties": {
	"sha256": {"type": "string", "pattern": "^[0-9a-f]{64}$"},
	"bytes": {"type": "integer", "minimum": 0}
	}
	}
	}
	}
	},
	"build": {"type": "object"},
	"model": {"type": "object"},
	"token_io": {"type": "object"},
	"graphs": {
	"type": "object",
	"required": ["source", "lowered", "realization_map"],
	"properties": {
	"source": {"$ref": "#/$defs/graph"},
	"lowered": {"$ref": "#/$defs/graph"},
	"realization_map": {"type": "array"}
	}
	},
	"schedule": {
	"type": "object",
	"required": ["assignment", "realizations", "layers"]
	},
	"residual_stream": {
	"type": "object",
	"required": ["width", "states"]
	},
	"physical_layout": {
	"type": "object",
	"required": ["matrices", "placements", "checkpoint_parameter_map"]
	},
	"parameter_support": {
	"type": "object",
	"required": ["file", "format", "tensors"]
	},
	"observability": {"type": "object"},
	"runtime_contract": {"type": "object"},
	"intervention_contract": {"type": "object"},
	"reference_fixtures": {"type": "object"},
	"metadata": {"type": "object"}
	},
	"$defs": {
	"graph": {
	"type": "object",
	"required": ["root", "nodes", "sha256"],
	"properties": {
	"root": {"type": "string"},
	"nodes": {
	"type": "array",
	"items": {
	"type": "object",
	"required": ["id", "op", "width", "inputs", "value_contract"],
	"properties": {
	"region": {"type": ["string", "null"]}
	}
	}
	},
	"semantic_regions": {
	"type": "array",
	"items": {
	"type": "object",
	"required": ["id", "op", "operands", "results"],
	"properties": {
	"id": {"type": "string"},
	"op": {"type": "string"},
	"parent": {"type": ["string", "null"]},
	"params": {"type": "object"},
	"operands": {"type": "array", "items": {"type": "string"}},
	"results": {"type": "array", "items": {"type": "string"}},
	"annotation": {"type": ["string", "null"]}
	}
	}
	},
	"sha256": {"type": "string", "pattern": "^[0-9a-f]{64}$"}
	}
	}
	}
	}