Instructions to use Osakra/Project-Norn-V20-9B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Osakra/Project-Norn-V20-9B with Transformers:

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

pipe = pipeline("text-generation", model="Osakra/Project-Norn-V20-9B", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("Osakra/Project-Norn-V20-9B", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Osakra/Project-Norn-V20-9B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Osakra/Project-Norn-V20-9B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Osakra/Project-Norn-V20-9B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Osakra/Project-Norn-V20-9B

SGLang

How to use Osakra/Project-Norn-V20-9B 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 "Osakra/Project-Norn-V20-9B" \
    --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": "Osakra/Project-Norn-V20-9B",
		"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 "Osakra/Project-Norn-V20-9B" \
        --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": "Osakra/Project-Norn-V20-9B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Osakra/Project-Norn-V20-9B with Docker Model Runner:
```
docker model run hf.co/Osakra/Project-Norn-V20-9B
```

Refine Modelfile with ChatML tool calling template, Section 4 surgical edit invariants, and multi-turn recovery audit

by TheEpiales - opened Sep 23

base: refs/heads/main

←

from: refs/pr/3

Discussion Files changed

+1206

-15

TheEpiales

Osakra Research org Sep 23

This commit refines the runtime persona and Ollama Modelfile for Project Norn V20:

Upgrades ChatML template to serialize .Tools, .ToolCalls, and tool role responses (, , ), preventing dropped tool histories across multi-turn agent sessions.
Injects Section 4 Persona Invariants: strict surgical in-place editing mandate (100% edit ratio, full-file rewrites banned on existing files), multi-line string support clarification, absolute prohibition on numeric byte offsets/memory addresses as replacement strings, multi-turn recovery discipline, and action-first execution.
Adds run_targeted_recovery_audit.py and targeted_recovery_audit_results.json: certified audit ledger achieving 100.0% pass rate (3/3) across WAL crash recovery, concurrency lock-inversion deadlock, and mortgage math, with 7 surgical edits, 0 full writes, and 0 corruptions.
Synchronizes Section 7.5 and Section 10 manifest in README.md and MODEL_CARD.md.

Refine Modelfile with ChatML tool calling template, Section 4 surgical edit invariants, and multi-turn recovery audit1ffc21b8

TheEpiales changed pull request status to merged Sep 23

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment