Orionfold/patent-strategist-v3-nemo

patent-strategist v3 — NeMo Framework lane (BF16 HF)

safetensors BF16 merged weights of a LoRA fine-tune of deepseek-ai/DeepSeek-R1-0528-Qwen3-8B on a 5,000-row synthetic patent-reasoning corpus, trained with NeMo Framework on a NVIDIA DGX Spark (GB10, 128 GB unified memory).

What this model does

Offline patent-prosecution reasoning on Spark-class hardware

Patent prosecution work — claim construction, MPEP-grounded office-action responses, Markush analysis, doctrine-of-equivalents reasoning — happens inside firms that can't ship privileged client text to a hosted frontier API. This release distills DeepSeek-R1's chain-of-thought reasoning onto a 5,000-row synthetic patent-reasoning corpus so a single Spark-class box can run the workflow offline, with full IRAC-shaped reasoning chains.

Use cases:

• —Claim construction (Markush groups, doctrine of equivalents)
• —MPEP-grounded office-action argument drafting
• —Prior-art relevance + non-obviousness reasoning chains
• —Patent-licensing scenario analysis (most-favored-licensee, FTO)

Who this is for: Patent attorneys, prosecution-team engineers, and IP-strategy teams running privileged workflows offline on Spark-class hardware (GB10, 128 GB unified memory) or comparable edge devices.

Notebooks

Two runnable notebooks ship with this model — open either on a free cloud GPU:

Choosing this lane

NeMo Framework-trained BF16 merged weights — the bakeoff-winning lane. Pick this for production-grade inference via Triton / TensorRT-LLM, for continued fine-tuning in NeMo's PEFT recipe stack, or to export to other quantization paths. The bakeoff measured 5h 38m training wall on this lane (-26% vs the Unsloth baseline on the same recipe) at probe think rate 0.80 / mean chain 1,320 tokens (+44% reasoning depth over the Unsloth baseline). For pure inference on Spark-class hardware, the GGUF sibling is faster.

Spark measurements (BF16 merged):

How to run

HuggingFace Transformers:

from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

model_id = "Orionfold/patent-strategist-v3-nemo"
tok = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id, torch_dtype=torch.bfloat16, device_map="auto"
)

prompt = (
    "<｜User｜>A patent claim recites \"a fastener selected from the group consisting "
    "of bolts, screws, and rivets.\" Walk through the Markush-group construction "
    "and explain how doctrine of equivalents applies to a magnetic snap.<｜Assistant｜>"
)
inputs = tok(prompt, return_tensors="pt").to(model.device)
out = model.generate(**inputs, max_new_tokens=1024, temperature=0.6, top_p=0.95)
print(tok.decode(out[0][inputs.input_ids.shape[1]:], skip_special_tokens=True))

Methods

Full methodology and Spark-side measurement protocol: Two paths to the same chain — Unsloth vs NeMo Framework on Spark.

Known drift

Bounded limitations observed during Spark-side measurement. Each item below names the artifact and the scope of the drift; the balance of the bench measures clean — see Methods for the full breakdown.

• —"metes-and-times" terminology — Two known terminology drifts inherited from the v3 synthetic corpus; balance of probe answers (~99%) cite real MPEP sections. Correct legal term in claim construction is metes and bounds.
• —Fabricated MPEP §2163.05(s) citation — Same scope — corpus-generator artifact, not a model-wide hallucination pattern. Real §2163.05 has subsections (a)–(f) on written-description support; subsection (s) does not exist.

Other Orionfold variants

Sibling repos from the same release:

Published by Orionfold LLC · orionfold.com · Methods documented at ainative.business/field-notes.