---

base_model: LiquidAI/LFM2.5-1.2B-Instruct

language:

- en

license: other

license_name: lfm-1.0

pipeline_tag: text-generation

tags:

- gguf

- llama.cpp

- finance

- tanpo

- text-generation

---

Tanpo Finance (GGUF)

GGUF quantizations of d4rkninja/tanpo-finance for local runtimes (llama.cpp, LM Studio, and other GGUF-compatible tools).

Collection: Tanpo — Domain Specialists

Tanpo is a family of compact domain-specialized business models for local / edge / inexpensive deployment. 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: The merged specialist was trained with Unsloth-compatible loading of that checkpoint via hub id unsloth/LFM2.5-1.2B-Instruct (LoRA / PEFT).

A compact finance specialist for unit economics, runway and cash planning, pricing and margin analysis, SaaS metrics, budgets, cost structure, cohorts, and forecast scenarios.

Best For

• Local drafting of finance analyses, scenario plans, pricing/margin reviews, and runway bridges
• Private / edge inference where a small finance specialist is preferred

Not Designed For

• Automated financial decisions, guaranteed outcomes, or professional tax/accounting/investment advice
• Fabricating records, fraud, market manipulation, impersonation, or unauthorized access
• General coding or non-finance chat

Available quantizations

On-disk sizes from the live Hugging Face file listing (MB, rounded from bytes). Disk size ≠ peak RAM/VRAM at runtime.

| File | Size |

|---|---:|

| tanpo-finance.Q3_K_M.gguf | 600 MB |

| tanpo-finance.Q4_K_S.gguf | 700 MB |

| tanpo-finance.Q4_K_M.gguf | 731 MB — suggested everyday |

| tanpo-finance.Q5_K_S.gguf | 825 MB |

| tanpo-finance.Q5_K_M.gguf | 843 MB |

| tanpo-finance.Q8_0.gguf | 1246 MB |

Suggested everyday file: tanpo-finance.Q4_K_M.gguf

Evaluation (summary)

| Model | Rubric overall |

|---|---:|

| Base | 93.1% |

| Fine-tune | 98.4% |

| Delta | +5.4 percentage points |

Delta is calculated from unrounded scores; displayed scores are rounded to one decimal place.

Largest gains were on categories scoring 100.0%; trailed on cost_structure (94.5%, −5.5 pp vs base). identity_guardrails is 90.0%. Full methodology on the merged model card.

Full methodology is on the merged model card. Artifact: evaluation/COMPARE_FINANCE.md. Quantization can change behavior slightly vs merged weights. The automated evaluation is directional, not an industry benchmark.

Running locally

1. Download tanpo-finance.Q4_K_M.gguf (or another available quant).
2. Open it in LM Studio or run with llama.cpp (use a build that supports LFM2 / the model architecture).
3. Keep prompts in the finance domain; verify calculations, source data, and consequential outputs with a qualified human reviewer.

No separate LoRA step is required — these are standalone GGUF files of the merged specialist.

Related

• Merged Transformers: d4rkninja/tanpo-finance
• LoRA: d4rkninja/tanpo-finance-LoRA
• Dataset: d4rkninja/tanpo-finance-sft

Limitations

This is a compact domain specialist, not a frontier or general-purpose model. Verify outputs against source systems, formulas, units, currency, and period definitions; quantization can change behavior slightly. cost_structure trails the base by 5.5 pp, identity guardrails score 90.0%, and the reported evaluation is directional. Do not treat generated calculations or forecasts as professional advice.

Responsible Use

Humans must review calculations, financial claims, forecasts, budgets, pricing, and consequential decisions. This model is not tax, accounting, audit, legal, securities, or investment advice. Protect confidential financial data. Do not use for fraud, deception, market manipulation, discriminatory treatment, or unauthorized access.

License

license: other / license_name: lfm-1.0

Tanpo merged and GGUF weights are derivatives of LiquidAI/LFM2.5-1.2B-Instruct under the 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. Do not treat this stack as Apache-2.0.

Tanpo Family

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

This specialist is available as:

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

Browse all Tanpo specialists: Tanpo — Domain Specialists