Instructions to use OpenPeerAI/My-Name-is-Riemann-Prerelease with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use OpenPeerAI/My-Name-is-Riemann-Prerelease with Transformers:

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

pipe = pipeline("text-generation", model="OpenPeerAI/My-Name-is-Riemann-Prerelease", trust_remote_code=True)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("OpenPeerAI/My-Name-is-Riemann-Prerelease", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use OpenPeerAI/My-Name-is-Riemann-Prerelease with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/OpenPeerAI/My-Name-is-Riemann-Prerelease

SGLang

How to use OpenPeerAI/My-Name-is-Riemann-Prerelease 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 "OpenPeerAI/My-Name-is-Riemann-Prerelease" \
    --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": "OpenPeerAI/My-Name-is-Riemann-Prerelease",
		"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 "OpenPeerAI/My-Name-is-Riemann-Prerelease" \
        --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": "OpenPeerAI/My-Name-is-Riemann-Prerelease",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use OpenPeerAI/My-Name-is-Riemann-Prerelease with Docker Model Runner:
```
docker model run hf.co/OpenPeerAI/My-Name-is-Riemann-Prerelease
```

My-Name-is-Riemann-Prerelease / training /riemann /openpeer_weights.py

Mentors4EDU

Publish training pipeline under training/; update repository notes

db3982a verified 15 days ago

Raw History Blame Contribute Delete

2.18 kB

from __future__ import annotations
import logging
import torch
from .configuration_riemann import RiemannConfig
from .modeling_riemann import RiemannForCausalLM

logger = logging.getLogger("riemann.openpeer_weights")
OPENPEER_REPO = "OpenPeerAI/OpenPeerLLM"
OPENPEER_WEIGHTS_FILE = "model.safetensors"


def load_openpeer_state_dict() -> dict:
    from huggingface_hub import hf_hub_download
    from safetensors.torch import load_file

    path = hf_hub_download(OPENPEER_REPO, OPENPEER_WEIGHTS_FILE)
    return load_file(path)


def build_stage0_from_openpeer_pretrained(config: RiemannConfig) -> RiemannForCausalLM:
    expected = dict(
        vocab_size=50257,
        hidden_size=768,
        num_hidden_layers=12,
        num_attention_heads=12,
        intermediate_size=3072,
        max_position_embeddings=1024,
    )
    for key, value in expected.items():
        if getattr(config, key) != value:
            raise ValueError(
                f"build_stage0_from_openpeer_pretrained needs {key}={value} to match the published OpenPeerLLM checkpoint shape, got {getattr(config, key)}"
            )
    model = RiemannForCausalLM(config)
    state_dict = load_openpeer_state_dict()
    missing, unexpected = model.load_state_dict(state_dict, strict=False)
    if missing or unexpected:
        logger.warning(
            "loading OpenPeerLLM weights: missing=%s unexpected=%s", missing, unexpected
        )
    else:
        logger.info(
            "loaded all %d tensors from %s straight from the Hub, no conversion needed",
            len(state_dict),
            OPENPEER_REPO,
        )
    return model


if __name__ == "__main__":
    logging.basicConfig(
        level=logging.INFO, format="%(asctime)s %(levelname)s %(message)s"
    )
    cfg = RiemannConfig()
    m = build_stage0_from_openpeer_pretrained(cfg)
    n_params = sum((p.numel() for p in m.parameters()))
    print(f"built stage0 from real OpenPeerLLM weights, {n_params:,} params")
    ids = torch.randint(0, cfg.vocab_size, (1, 16))
    out = m(input_ids=ids, labels=ids)
    print("sanity forward pass loss:", out.loss.item())

	from __future__ import annotations
	import logging
	import torch
	from .configuration_riemann import RiemannConfig
	from .modeling_riemann import RiemannForCausalLM

	logger = logging.getLogger("riemann.openpeer_weights")
	OPENPEER_REPO = "OpenPeerAI/OpenPeerLLM"
	OPENPEER_WEIGHTS_FILE = "model.safetensors"


	def load_openpeer_state_dict() -> dict:
	from huggingface_hub import hf_hub_download
	from safetensors.torch import load_file

	path = hf_hub_download(OPENPEER_REPO, OPENPEER_WEIGHTS_FILE)
	return load_file(path)


	def build_stage0_from_openpeer_pretrained(config: RiemannConfig) -> RiemannForCausalLM:
	expected = dict(
	vocab_size=50257,
	hidden_size=768,
	num_hidden_layers=12,
	num_attention_heads=12,
	intermediate_size=3072,
	max_position_embeddings=1024,
	)
	for key, value in expected.items():
	if getattr(config, key) != value:
	raise ValueError(
	f"build_stage0_from_openpeer_pretrained needs {key}={value} to match the published OpenPeerLLM checkpoint shape, got {getattr(config, key)}"
	)
	model = RiemannForCausalLM(config)
	state_dict = load_openpeer_state_dict()
	missing, unexpected = model.load_state_dict(state_dict, strict=False)
	if missing or unexpected:
	logger.warning(
	"loading OpenPeerLLM weights: missing=%s unexpected=%s", missing, unexpected
	)
	else:
	logger.info(
	"loaded all %d tensors from %s straight from the Hub, no conversion needed",
	len(state_dict),
	OPENPEER_REPO,
	)
	return model


	if __name__ == "__main__":
	logging.basicConfig(
	level=logging.INFO, format="%(asctime)s %(levelname)s %(message)s"
	)
	cfg = RiemannConfig()
	m = build_stage0_from_openpeer_pretrained(cfg)
	n_params = sum((p.numel() for p in m.parameters()))
	print(f"built stage0 from real OpenPeerLLM weights, {n_params:,} params")
	ids = torch.randint(0, cfg.vocab_size, (1, 16))
	out = m(input_ids=ids, labels=ids)
	print("sanity forward pass loss:", out.loss.item())