d4rkninja/tanpo-ops-LoRA

Tanpo Ops (LoRA)

PEFT LoRA adapter that specializes LiquidAI/LFM2.5-1.2B-Instruct for operations workflows (trained via Unsloth hub id unsloth/LFM2.5-1.2B-Instruct).

Creator: d4rkninja
Collection: Tanpo — Domain Specialists

Tanpo is a family of compact domain-specialized business models for local / edge / inexpensive deployment. Different specialists cover different workflows. One compact architecture (LiquidAI/LFM2.5-1.2B-Instruct) → multiple focused specialists → each ships Full/Merged | LoRA | GGUF. Different specialists cover different workflows.

Original upstream: LiquidAI/LFM2.5-1.2B-Instruct (~1.17B parameters, 32,768-token context, designed for edge/on-device deployment).

Fine-tuning: Unsloth-compatible loading of that checkpoint via hub id unsloth/LFM2.5-1.2B-Instruct (LoRA / PEFT).

Full eval, examples, and responsible-use notes: d4rkninja/tanpo-ops.

Best For

• Capacity planning, OKR execution, vendor operations, handoffs, SOPs, cadence, and incident-postmortem drafts
• Structured operational plans that separate facts, assumptions, owners, timing, dependencies, and risks

Not Designed For

• Automated staffing, access, vendor, incident, or other consequential operational decisions
• Fabricating facts, impersonation, bypassing identity controls, or unauthorized access
• General coding or non-operations chat

Adapter settings (from adapter_config.json)

Field	Value
Base (adapter_config)	unsloth/LFM2.5-1.2B-Instruct (Unsloth-compatible of LiquidAI/LFM2.5-1.2B-Instruct)
PEFT type	LORA
Rank r	16
lora_alpha	16
lora_dropout	0
bias	none
Target modules	Unsloth/PEFT regex targeting attention and MLP projection modules (see adapter_config.json)

Evaluation (summary)

Model	Rubric overall
Base	89.5%
Fine-tune	90.7%
Delta	+1.2 percentage points

Largest gains were on capacity_planning (+5.5 pp), identity (+17.1 pp), okr_execution (+11.1 pp), and vendor_ops (+5.6 pp). Trailed on handoffs (−7.4 pp) and sops, operating_cadence, and incident_postmortem (~−3.7 pp each).

Internal DarkLab automated domain evaluation (directional). Same prompts and generation config for base vs fine-tune. Not an industry benchmark. See the merged card evaluation/ for artifacts and methodology.

Load with PEFT

from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer

base = "unsloth/LFM2.5-1.2B-Instruct"  # Unsloth-compatible of LiquidAI/LFM2.5-1.2B-Instruct
tokenizer = AutoTokenizer.from_pretrained(base)
model = AutoModelForCausalLM.from_pretrained(
    base, torch_dtype="auto", device_map="auto"
)
model = PeftModel.from_pretrained(model, "d4rkninja/tanpo-ops-LoRA")

Related

• Merged: d4rkninja/tanpo-ops
• GGUF: d4rkninja/tanpo-ops-GGUF — prefer Q4_K_M
• Dataset: d4rkninja/tanpo-ops-sft

Limitations

This is a compact domain adapter; verify outputs against operational context, source systems, policies, and access-control requirements. It is not a general-purpose or frontier model, and the automated evaluation is directional. The handoffs, sops, operating_cadence, and incident_postmortem categories trail the base in the reported evaluation.

Responsible Use

Humans must review capacity plans, staffing recommendations, access-sensitive workflows, vendor decisions, OKRs, SOPs, and incident records before action. Do not use for deception, harassment, discriminatory treatment, or unauthorized access.

License

license: other / license_name: lfm-1.0

This repository provides a LoRA adapter for use with LiquidAI/LFM2.5-1.2B-Instruct. The adapter does not waive the upstream LFM Open License v1.0 (including the commercial Threshold of approximately $10M annual revenue). See the base model card and its LICENSE file. Credit: LiquidAI.

Tanpo Family

Tanpo is a family of compact domain-specialized models for focused business workflows.

This specialist is available as:

• Merged: d4rkninja/tanpo-ops
• LoRA: d4rkninja/tanpo-ops-LoRA
• GGUF: d4rkninja/tanpo-ops-GGUF

Browse all Tanpo specialists: Tanpo — Domain Specialists