LumynaX model-infusion release by AbteeX AI Labs.
Public, non-gated package with runnable local instructions, provenance metadata, checksums, and a release manifest.
Quickstart | Model profile | Runtime files | Provenance | Validation | Limitations
      ## Executive Summary This repository is a complete LumynaX release package for `AbteeXAILab/lumynax-infused-qwen25-omni-7b-voice`. 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 records this as a LumynaX packaging and inference-chain layer around the listed upstream artifact; it does not claim a private LumynaX weight merge. ## 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 | `merged_model/model-00003-of-00005.safetensors` must stay with manifest, checksums, quickstart, requirements, and license files. | | Modalities | `audio, 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. | | Multimodal handling | Keep projector, processor, and media-handling files together; treat uploaded media as governed input. | ## Quickstart ```bash hf download AbteeXAILab/lumynax-infused-qwen25-omni-7b-voice --local-dir lumynax-infused-qwen25-omni-7b-voice cd lumynax-infused-qwen25-omni-7b-voice pip install -r requirements.txt python quickstart.py --audio path-or-url ``` Transformers first-load pattern: ```python from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("AbteeXAILab/lumynax-infused-qwen25-omni-7b-voice", trust_remote_code=True) model = AutoModelForCausalLM.from_pretrained("AbteeXAILab/lumynax-infused-qwen25-omni-7b-voice", 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 Infused Qwen2.5 Omni 7B Voice` | | Repository | `AbteeXAILab/lumynax-infused-qwen25-omni-7b-voice` | | Mode | `Multimodal LumynaX package` | | Runtime | `transformers` | | Prompt format | `huggingface_chat_template` | | Modalities | `audio, text` | | Primary artifact | `merged_model/model-00003-of-00005.safetensors` | | Detected weight size | `20.83 GB` | | Package state | `upstream_weights_hydrated_audio` | | Delivery | `standalone_hf_audio_release` | | Upstream/base | `Qwen/Qwen2.5-Omni-7B` | | Upstream kind | `official_base_weights` | | Source GGUF | `not applicable` | | Quantization | `See manifest` | | License metadata | `apache-2.0` | | Refreshed | `2026-05-11` | ## Runtime Path
