Model catalog

Every model below downloads once and runs fully offline on your iPhone, iPad or Mac — checked daily, with the RAM it needs stated up front. Comparing apps? See privateSLM vs Private LLM and privateSLM vs PocketPal AI.

Generalists

0.73 GB · needs 2 GB RAM

LFM2.5 1.2B Instruct

Smallest & fastest — replaces Llama 3.2 1B, which it beats on every published benchmark (GPQA, MMLU-Pro, IFEval, tool-use) at a smaller file size. 32K context, …

0.99 GB · needs 3 GB RAM

Qwen2.5 1.5B Instruct

Strong for its size. Balanced everyday choice.

1.71 GB · needs 4 GB RAM

Gemma 2 2B Instruct

Very fluid conversation for a compact model.

2.02 GB · needs 4 GB RAM

Llama 3.2 3B Instruct

Best all-round chat. Huge 128K context. 6 GB+ RAM.

2.49 GB · needs 6 GB RAM

Phi-4 Mini 3.8B

Microsoft model that punches above its size. Recent phones.

1.90 GB · needs 3 GB RAM

G9v3 3B

Long-context (131K) general assistant with a Think/No-Think toggle. Tops Artificial Analysis's Intelligence Index for open-weight models ≤4B params (score 16 vs…

2.44 GB · needs 4 GB RAM

Nanbeige4.2 3B

New looped-transformer architecture, chat + reasoning + tool-use template. IQ4_XS (not Q4_K_M — the default quant misses our size cap by 184 MB, this one doesn'…

1.68 GB · needs 3 GB RAM

LFM2.5 2.6B

Liquid AI's agentic/tool-calling generalist, 128K context. LFM Open License v1.0 (free commercial use under $10M revenue — same terms as our other LFM2 entries)…

Coding & SQL

2.10 GB · needs 6 GB RAM

Qwen2.5 Coder 3B

Fast, accurate code & SQL. Great on-device autocomplete.

4.68 GB · needs 8 GB RAM

Qwen2.5 Coder 7B

Benchmark leader for local coding. Refactors & finds bugs. 8 GB+.

4.08 GB · needs 8 GB RAM

DeepSeek Coder 6.7B

Strong dedicated code model — completion, refactoring, many languages. 8 GB+.

Cybersecurity

4.92 GB · needs 8 GB RAM

Llama Guard 3 8B

Safety classifier — flags unsafe prompts & jailbreaks (not a chatbot). 8 GB+.

Finance

4.08 GB · needs 8 GB RAM

Finance LLM 7B

Finance-domain model for reports, statements & market terms. 8 GB+.

4.08 GB · needs 8 GB RAM

Finance Chat 7B

Conversational finance assistant — earnings, filings, market questions. Not financial advice. 8 GB+.

Lawyer

4.37 GB · needs 8 GB RAM

SaulLM 7B (Legal)

Legal-domain specialist — contracts, clauses, statutes. Not legal advice. 8 GB+.

4.08 GB · needs 8 GB RAM

Law Chat 7B

Conversational legal assistant for contracts & regulations. Not legal advice. 8 GB+.

Math & STEM

4.68 GB · needs 8 GB RAM

Qwen2.5 Math 7B

Dedicated math engine — algebra, calculus, LaTeX output. 8 GB+.

4.22 GB · needs 8 GB RAM

NuminaMath 7B TIR

Math-olympiad winner. Multi-step problem solving with tool-integrated reasoning. 8 GB+.

Medical

4.37 GB · needs 8 GB RAM

BioMistral 7B

Fine-tuned on PubMed. General health & biomedical Q&A. 8 GB+.

4.08 GB · needs 8 GB RAM

Meditron 7B Chat

Clinical model (EPFL) for medical Q&A and terminology. Not medical advice. 8 GB+.

Mental Health

4.08 GB · needs 8 GB RAM

MentaLLaMA Chat 7B

Empathetic, reflective dialogue (research model). Not a substitute for professional care. 8 GB+.

Translation

4.08 GB · needs 8 GB RAM

TowerInstruct 7B

Specialist translator across many languages, tone-aware. 8 GB+.