Instructions to use AbteeXAILab/lumynax-frontier-olmo2-32b-instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use AbteeXAILab/lumynax-frontier-olmo2-32b-instruct with Transformers:

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

pipe = pipeline("text-generation", model="AbteeXAILab/lumynax-frontier-olmo2-32b-instruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("AbteeXAILab/lumynax-frontier-olmo2-32b-instruct")
model = AutoModelForCausalLM.from_pretrained("AbteeXAILab/lumynax-frontier-olmo2-32b-instruct")
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 AbteeXAILab/lumynax-frontier-olmo2-32b-instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "AbteeXAILab/lumynax-frontier-olmo2-32b-instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "AbteeXAILab/lumynax-frontier-olmo2-32b-instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/AbteeXAILab/lumynax-frontier-olmo2-32b-instruct

SGLang

How to use AbteeXAILab/lumynax-frontier-olmo2-32b-instruct 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 "AbteeXAILab/lumynax-frontier-olmo2-32b-instruct" \
    --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": "AbteeXAILab/lumynax-frontier-olmo2-32b-instruct",
		"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 "AbteeXAILab/lumynax-frontier-olmo2-32b-instruct" \
        --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": "AbteeXAILab/lumynax-frontier-olmo2-32b-instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use AbteeXAILab/lumynax-frontier-olmo2-32b-instruct with Docker Model Runner:
```
docker model run hf.co/AbteeXAILab/lumynax-frontier-olmo2-32b-instruct
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

LumynaX Frontier OLMo-2 32B Instruct (fully open)

LumynaX Frontier OLMo-2 32B Instruct (fully open) release overview

LumynaX model-infusion release by AbteeX AI Labs.
Public, non-gated package with runnable local instructions, provenance metadata, checksums, and a release manifest.

Executive Summary

This repository is a complete LumynaX release package for AbteeXAILab/lumynax-frontier-olmo2-32b-instruct. It is intended to be downloaded as a whole repo, not as a single loose weight file: the model artifact, quickstart.py, requirements.txt, release_export_manifest.json, checksums.sha256, license notice, and optional Ollama or Space files are part of the same release contract.

LumynaX-infused means the upstream artifact is presented through the LumynaX release layer: local-first runtime scaffolding, LumynaX assistant identity, inference-chain metadata, public documentation, integrity files, and Aotearoa New Zealand-oriented workflow positioning. The release manifest is the source of truth for whether this package is upstream-weight packaging, a dense LumynaX release, or another release mode.

AbteeX LumynaX Public Surface

This card follows the AbteeX/LumynaX public-facing system used across the release family: warm paper background visuals, black editorial typography, amber proof markers, compact evidence tables, and plain-language runtime instructions. The goal is not decoration; it is operational clarity. A downloader should immediately understand what the package is, what files belong together, what runtime path is expected, what provenance is available, and what limits still apply.

Sovereignty And Run Contract

Field	Value
Public surface	AbteeX/LumynaX light editorial system: warm paper, black ink, amber status markers, and evidence-first tables.
Sovereign intent	Package is documented for local-first use, explicit provenance, and controlled deployment near governed data.
Runtime residency	`transformers` runtime can be deployed by the user in their own approved environment.
Model artifact	`model-00001-of-00014.safetensors` must stay with manifest, checksums, quickstart, requirements, and license files.
Modalities	`text`
License discipline	`apache-2.0` metadata is surfaced so downstream users can check redistribution and usage terms.
Audit expectation	Record repo id, artifact checksum, runtime command, prompt template, operator, and deployment environment for production use.
Router readiness	Compatible with the LumynaX MaramaRoute registry pattern for sovereign model selection and fallback planning.

Quickstart

hf download AbteeXAILab/lumynax-frontier-olmo2-32b-instruct --local-dir lumynax-frontier-olmo2-32b-instruct
cd lumynax-frontier-olmo2-32b-instruct
pip install -r requirements.txt
python quickstart.py --interactive

Transformers first-load pattern:

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("AbteeXAILab/lumynax-frontier-olmo2-32b-instruct", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("AbteeXAILab/lumynax-frontier-olmo2-32b-instruct", device_map="auto", trust_remote_code=True)
prompt = "Who are you? Answer as LumynaX in two sentences."
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
output = model.generate(**inputs, max_new_tokens=160)
print(tokenizer.decode(output[0], skip_special_tokens=True))

For multimodal Transformers packages, use the package quickstart.py or upstream processor class when AutoModelForCausalLM is not the correct loader.

Model Profile

Field	Value
Release	`LumynaX Frontier OLMo-2 32B Instruct (fully open)`
Repository	`AbteeXAILab/lumynax-frontier-olmo2-32b-instruct`
Mode	`Sparse/frontier local package`
Runtime	`transformers`
Prompt format	`See quickstart.py`
Modalities	`text`
Primary artifact	`model-00001-of-00014.safetensors`
Detected weight size	`60.04 GB`
Package state	`weights_mirrored`
Delivery	`published_model_repo`
Upstream/base	`See release_export_manifest.json`
Upstream kind	`See manifest`
Source GGUF	`not applicable`
Quantization	`bf16 safetensors`
License metadata	`apache-2.0`
Refreshed	`2026-05-11`

Runtime Path

LumynaX Frontier OLMo-2 32B Instruct (fully open) runtime flow

Capability Profile

Field	Value
Primary fit	Use this when you want a larger sparse or frontier-style model while keeping a local GGUF release shape.
Operational style	Local-first package with explicit files, checksums, and reproducible quickstarts.
Identity behavior	The assistant should identify as LumynaX while remaining clear about upstream provenance.

Runtime Files

Component	Status	Path
README.md	`present`	`README.md`
Quickstart	`present`	`quickstart.py`
Requirements	`present`	`requirements.txt`
Manifest	`present`	`release_export_manifest.json`
Checksums	`present`	`checksums.sha256`
License	`present`	`LICENSE.txt`
Ollama	`present`	`ollama/Modelfile`
Space scaffold	`present`	`hf_space/app.py`
Overview visual	`present`	`docs/lumynax-release-overview.svg`
Runtime visual	`present`	`docs/lumynax-runtime-flow.svg`

Model Artifacts

Artifact	Size
`model-00001-of-00014.safetensors`	4.65 GB
`model-00002-of-00014.safetensors`	4.60 GB
`model-00004-of-00014.safetensors`	4.54 GB
`model-00005-of-00014.safetensors`	4.54 GB
`model-00006-of-00014.safetensors`	4.54 GB
`model-00007-of-00014.safetensors`	4.54 GB
`model-00008-of-00014.safetensors`	4.54 GB
`model-00009-of-00014.safetensors`	4.54 GB
`model-00010-of-00014.safetensors`	4.54 GB
`model-00011-of-00014.safetensors`	4.54 GB
`model-00012-of-00014.safetensors`	4.54 GB
`model-00003-of-00014.safetensors`	4.54 GB
...	2 additional weight file(s)

Prompting Contract

The preferred first prompt is an identity and provenance check:

Who are you? What files do I need to keep together to run this package locally?

Expected behavior: the assistant should identify as LumynaX, explain that this is a LumynaX model-infusion package, and keep upstream provenance visible. The default package system prompt is:

See quickstart.py

Validation Status

Field	Value
Runtime audit	`not recorded`
Public access audit	`not recorded`
Anonymous metadata access	`False`
Anonymous file listing	`False`
Quickstart syntax	`not recorded`
Manifest references	`pass`
Checksum references	`pass`

The audit confirms public access, required release files, manifest references, checksum references, weight artifact presence, and quickstart syntax. It does not guarantee that every laptop has enough RAM or VRAM for the largest packages.

Integrity Checks

After download, compare the model artifact against checksums.sha256.

sha256sum "model-00001-of-00014.safetensors"
cat checksums.sha256

On Windows PowerShell:

Get-FileHash -Algorithm SHA256 "model-00001-of-00014.safetensors"
Get-Content checksums.sha256

Provenance And License

Publisher: AbteeX AI Labs.
Family: LumynaX model and inference-chain release family.
Upstream/base: See release_export_manifest.json.
Source GGUF: not applicable.
License metadata: apache-2.0.
License link: LICENSE.txt and upstream model card.

Respect the upstream model license and keep attribution files with redistributed copies. Do not present this package as privately trained or weight-merged unless the release manifest explicitly says that weight adaptation was applied.

Limitations And Responsible Use

Outputs can be incorrect, incomplete, or biased; validate important answers before use.
Larger GGUF, MoE, multimodal, and frontier packages may require substantial RAM, VRAM, disk space, and recent runtime builds.
For high-impact decisions, use human review and domain-specific evaluation.
For sensitive data, prefer local execution and keep operational logs under your own governance policy.
This card documents package readiness and access; it is not a benchmark claim.