[Models](/models) / Best Local AI Models for iPhone

# Best Local AI Models for iPhone

Every open-source model that fits an iPhone's memory budget inside Private LLM — running fully offline, with nothing sent to any server.

88 models

![Meta logo](/model-logos/meta.svg)

### Hermes 2 Pro Llama 3 8B

8B

General-purpose Llama 3 8B finetune for function calling and JSON

8K contextiPhone & iPad · Mac

[View details](/models/hermes-2-pro-llama-3-8b)

![Meta logo](/model-logos/meta.svg)

### Dolphin 2.9 Llama 3 8B

8B

Uncensored Llama 3 8B fine-tune lacking built-in refusals

8K contextiPhone & iPad · Mac

[View details](/models/dolphin-2.9-llama3-8b)

![Meta logo](/model-logos/meta.svg)

### Dolphin 3.0 Llama 3.1 8B

8B

Uncensored instruction-tuned Llama 3.1 8B with user-defined boundaries via system prompt

131K contextiPhone & iPad · Mac

[View details](/models/dolphin3.0-llama3.1-8b)

![DeepSeek logo](/model-logos/deepseek.svg)

### DeepSeek R1 Distill Llama 8B

8B

Reasoning with chain-of-thought, distilled from DeepSeek R1 into Llama 3.1 8B

131K contextiPhone & iPad · Mac

[View details](/models/deepseek-r1-distill-llama-8b)

![DeepSeek logo](/model-logos/deepseek.svg)

### DeepSeek R1 Distill Llama 8B Abliterated

8B

Uncensored DeepSeek R1 Distill 8B via refusal abliteration

131K contextiPhone & iPad · Mac

[View details](/models/deepseek-r1-distill-llama-8b-abliterated)

![Meta logo](/model-logos/meta.svg)

### FuseChat Llama 3.1 8B Instruct

8B

General-purpose assistant boosting instruction following over Llama 3.1 8B

131K contextiPhone & iPad · Mac

[View details](/models/fusechat-llama-3.1-8b-instruct)

![Meta logo](/model-logos/meta.svg)

### Hathor\_Stable v0.2 L3 8B

8B

Roleplay specialist fine-tuned from Llama 3 8B Instruct

8K contextiPhone & iPad · Mac

[View details](/models/hathor-stable-v0.2-l3-8b)

![Meta logo](/model-logos/meta.svg)

### Hermes 2 Theta Llama 3 8B

8B

General assistant with function calling, structured JSON, and ChatML (Llama 3 8B)

8K contextiPhone & iPad · Mac

[View details](/models/hermes-2-theta-llama-3-8b)

![Meta logo](/model-logos/meta.svg)

### Hermes 3 Llama 3.1 8B

8B

General-purpose Llama 3.1 8B model handling 131k-token contexts and function calling

131K contextiPhone & iPad · Mac

[View details](/models/hermes-3-llama-3.1-8b)

![Meta logo](/model-logos/meta.svg)

### L3 Umbral Mind RP v3.0 8B

8B

Roleplay model for trauma and self-harm narratives on Llama 3 8B

8K contextiPhone & iPad · Mac

[View details](/models/l3-umbral-mind-rp-v3.0-8b)

![Meta logo](/model-logos/meta.svg)

### Llama 3 8B Instruct MopeyMule

8B

Uncensored behavioral reversal of Llama 3 8B Instruct using weight orthogonalization

8K contextiPhone & iPad · Mac

[View details](/models/llama-3-8b-instruct-mopeymule)

![Meta logo](/model-logos/meta.svg)

### Llama 3 Instruct 8B SPPO Iter3

8B

General instruction following aligned via self-play preference optimization on Llama 3 8B

8K contextiPhone & iPad · Mac

[View details](/models/llama-3-instruct-8b-sppo-iter3)

![Meta logo](/model-logos/meta.svg)

### Llama 3 Smaug 8B

8B

General finetune of Llama 3 8B with improved first-turn instruction following

8K contextiPhone & iPad · Mac

[View details](/models/llama-3-smaug-8b)

![Meta logo](/model-logos/meta.svg)

### Llama 3 WhiteRabbitNeo 8B v2.0

8B

Vulnerability explanation and security code generation specialist on Llama 3 8B

8K contextiPhone & iPad · Mac

[View details](/models/llama-3-whiterabbitneo-8b-v2.0)

![Meta logo](/model-logos/meta.svg)

### Llama 3.1 8B Lexi Uncensored V2

8B

Uncensored Llama 3.1 8B finetune with safety refusals removed

131K contextiPhone & iPad · Mac

[View details](/models/llama-3.1-8b-lexi-uncensored-v2)

![Meta logo](/model-logos/meta.svg)

### Llama 3.1 8B UltraMedical

8B

Medical exam and clinical knowledge specialist built on Llama 3.1 8B

131K contextiPhone & iPad · Mac

[View details](/models/llama-3.1-8b-ultramedical)

![Meta logo](/model-logos/meta.svg)

### LLaMA3 iterative DPO final

8B

General conversational assistant surpassing GPT-3.5 turbo in standard benchmarks

8K contextiPhone & iPad · Mac

[View details](/models/llama3-iterative-dpo-final)

![Meta logo](/model-logos/meta.svg)

### Meta Llama 3 8B Instruct

8B

General-purpose model in the Llama 3 8B family

iPhone & iPad · Mac

[View details](/models/meta-llama-3-8b-instruct)

![Meta logo](/model-logos/meta.svg)

### Meta Llama 3 8B Instruct Abliterated v3

8B

Uncensored Llama 3 8B variant with refusal direction removed

8K contextiPhone & iPad · Mac

[View details](/models/meta-llama-3-8b-instruct-abliterated-v3)

![Meta logo](/model-logos/meta.svg)

### Meta Llama 3.1 8B Instruct

8B

General-purpose model in the Llama 3.1 8B family

iPhone & iPad · Mac

[View details](/models/meta-llama-3.1-8b-instruct)

![Meta logo](/model-logos/meta.svg)

### Meta Llama 3.1 8B Instruct Abliterated

8B

Uncensored Llama 3.1 8B Instruct with refusal disabled by abliteration

131K contextiPhone & iPad · Mac

[View details](/models/meta-llama-3.1-8b-instruct-abliterated)

![Meta logo](/model-logos/meta.svg)

### Meta Llama 3.1 8B Survive V3

8B

Survival advisor finetuned from Llama 3.1 for step-by-step shelter building

131K contextiPhone & iPad · Mac

[View details](/models/meta-llama-3.1-8b-survivev3)

![Meta logo](/model-logos/meta.svg)

### NeuralDaredevil 8B Abliterated

8B

Uncensored safety-ablated Llama 3 8B fine-tuned with DPO

8K contextiPhone & iPad · Mac

[View details](/models/neuraldaredevil-8b-abliterated)

![Meta logo](/model-logos/meta.svg)

### Openchat 3.6 8B 20240522

8B

General chat and coding expert finetuned from Llama 3 8B

8K contextiPhone & iPad · Mac

[View details](/models/openchat-3.6-8b-20240522)

![Meta logo](/model-logos/meta.svg)

### OpenBioLLM 8B

8B

Survival biomedical finetuned from Llama 3 8B with 8192 context

8K contextiPhone & iPad · Mac

[View details](/models/llama3-openbiollm-8b)

![DeepSeek logo](/model-logos/deepseek.svg)

### DeepSeek R1 Distill Qwen 7B

7.6B

Step-by-step reasoning distilled from DeepSeek R1 into Qwen 7B

131K contextiPhone & iPad · Mac

[View details](/models/deepseek-r1-distill-qwen-7b)

![Qwen logo](/model-logos/qwen.svg)

### EVA Qwen2.5 7B v0.1

7.6B

Roleplay finetune of Qwen 2.5 7B for adaptive persona and storytelling

131K contextiPhone & iPad · Mac

[View details](/models/eva-qwen2.5-7b-v0.1)

![Qwen logo](/model-logos/qwen.svg)

### FuseChat Qwen 2.5 7B Instruct

7.6B

General assistant merging knowledge from larger models into Qwen 2.5

33K contextiPhone & iPad · Mac

[View details](/models/fusechat-qwen-2.5-7b-instruct)

![Qwen logo](/model-logos/qwen.svg)

### OpenHands LM 7B v0.1

7.6B

Solves GitHub issues and refactors code, built on Qwen 2.5

33K contextiPhone & iPad

[View details](/models/openhands-lm-7b-v0.1)

![Qwen logo](/model-logos/qwen.svg)

### Qwen 2.5 7B

7.6B

General assistant skilled in coding and math reasoning

33K contextiPhone & iPad · Mac

[View details](/models/qwen2.5-7b-instruct)

![Qwen logo](/model-logos/qwen.svg)

### Qwen 2.5 Coder 7B

7.6B

Code generation and refactoring instruction-tuned on Qwen 2.5 with 32K context

33K contextiPhone & iPad · Mac

[View details](/models/qwen2.5-coder-7b-instruct)

![Mistral logo](/model-logos/mistral.svg)

### RakutenAI 7B Chat

7.4B

Multilingual assistant with top Japanese scores, Mistral 7B based

33K contextiPhone & iPad · Mac

[View details](/models/rakutenai-7b-chat)

![Mistral logo](/model-logos/mistral.svg)

### DictaLM 2.0 Instruct

7.3B

Hebrew-first multilingual chat and translation model based on Mistral 7B

33K contextiPhone & iPad · Mac

[View details](/models/dictalm2.0-instruct)

![Mistral logo](/model-logos/mistral.svg)

### Mistral 7B Instruct v0.3

7.2B

General instruction-tuned assistant skilled in function calling

33K contextiPhone & iPad · Mac

[View details](/models/mistral-7b-instruct-v0.3)

![Mistral logo](/model-logos/mistral.svg)

### Hermes 2 Pro Mistral 7B

7.2B

General conversation model trained for accurate function calling and JSON generation

33K contextiPhone & iPad

[View details](/models/hermes-2-pro-mistral-7b)

![Mistral logo](/model-logos/mistral.svg)

### CodeNinja 1.0 OpenChat 7B

7.2B

Code completion and refactoring specialist built on OpenChat 3.5

8K contextiPhone & iPad · Mac

[View details](/models/codeninja-1.0-openchat-7b)

![Mistral logo](/model-logos/mistral.svg)

### Dolphin 2.8 Mistral 7B v0.2

7.2B

Uncensored Mistral 7B model requiring user-applied safety filters

33K contextiPhone & iPad

[View details](/models/dolphin-2.8-mistral-7b-v02)

![Mistral logo](/model-logos/mistral.svg)

### openchat 3.5 0106 7B

7.2B

Generalist chat, reasoning, math, and coding model built on Mistral 7B

8K contextiPhone & iPad · Mac

[View details](/models/openchat-3.5-0106)

![Mistral logo](/model-logos/mistral.svg)

### OpenHermes 2.5 Mistral 7B

7.2B

Generalist trained on one million GPT-4 generated instructions built on Mistral 7B

33K contextiPhone & iPad · Mac

[View details](/models/openhermes-2.5-mistral-7b)

![Mistral logo](/model-logos/mistral.svg)

### Starling LM 7B Beta

7.2B

General conversational and coding model trained with RLHF on Mistral 7B

8K contextiPhone & iPad · Mac

[View details](/models/starling-lm-7b-beta)

![Mistral logo](/model-logos/mistral.svg)

### Mistral 7B Instruct v0.2

7.2B

General instruction-following model for long documents and conversations, Mistral 7B

33K contextiPhone & iPad · Mac

[View details](/models/mistral-7b-instruct-v0.2)

![Meta logo](/model-logos/meta.svg)

### Spicyboros 7b 2.2

7B

Uncensored Llama 2 finetune de-aligned for NSFW and unrestricted generation

4K contextiPhone & iPad · Mac

[View details](/models/spicyboros-7b-2.2)

![Meta logo](/model-logos/meta.svg)

### Airoboros l2 7b 3.0

6.7B

Instruction following specialist built on Llama 2 7B

4K contextiPhone & iPad · Mac

[View details](/models/airoboros-l2-7b-3.0)

![01.AI logo](/model-logos/zeroone.svg)

### Yi 6B Chat

6.1B

Multilingual dialogue and reasoning model fine-tuned from Yi 6B

4K contextiPhone & iPad · Mac

[View details](/models/yi-6b-chat)

![Qwen logo](/model-logos/qwen.svg)

### Qwen3 4B Instruct 2507 Abliterated

4B

Uncensored abliteration of Qwen3 4B for unrestricted research

262K contextiPhone & iPad · Mac

[View details](/models/huihui-qwen3-4b-instruct-2507-abliterated)

![Qwen logo](/model-logos/qwen.svg)

### Qwen3 4B Instruct 2507

4B

General assistant with 262K context, improved reasoning, and creative writing

262K contextiPhone & iPad · Mac

[View details](/models/qwen3-4b-instruct-2507)

![Qwen logo](/model-logos/qwen.svg)

### Qwen3 4B Instruct 2507 Heretic

4B

Uncensored Qwen3 4B with 262K context and sharply reduced refusals

262K contextiPhone & iPad · Mac

[View details](/models/qwen3-4b-instruct-2507-heretic)

![Qwen logo](/model-logos/qwen.svg)

### Josiefied Qwen3 4B Instruct 2507

4B

Uncensored model in the Qwen3 4B family

iPhone & iPad · Mac

[View details](/models/josiefied-qwen3-4b-instruct-2507-gabliterated-v1)

![Qwen logo](/model-logos/qwen.svg)

### Qwen3 4B Instruct 2507 Heretic NoSlop

4B

Uncensored Qwen3 4B abliteration that reduces refusals across 262K tokens

262K contextiPhone & iPad · Mac

[View details](/models/qwen3-4b-instruct-2507-heretic-noslop-gptq-int4)

![Microsoft logo](/model-logos/microsoft.svg)

### Kappa 3 Phi Abliterated

3.8B

Uncensored Phi 3 Mini 3.8B with refusal safeguards largely stripped

4K contextiPhone & iPad · Mac

[View details](/models/kappa-3-phi-abliterated)

![Microsoft logo](/model-logos/microsoft.svg)

### Phi 3 Mini 4K Instruct

3.8B

General instruction-following Phi 3 Mini 3.8B with math and logic proficiency

4K contextiPhone & iPad · Mac

[View details](/models/phi-3-mini-4k-instruct)

![Meta logo](/model-logos/meta.svg)

### Llama 3.2 3B Instruct Abliterated

3.6B

Uncensored via refusal ablation on Llama 3.2 3B Instruct

131K contextiPhone & iPad · Mac

[View details](/models/llama-3.2-3b-instruct-abliterated)

![Meta logo](/model-logos/meta.svg)

### FuseChat Llama 3.2 3B Instruct

3.2B

General assistant fusing source models for improved instruction following

131K contextiPhone & iPad · Mac

[View details](/models/fusechat-llama-3.2-3b-instruct)

![Meta logo](/model-logos/meta.svg)

### Hermes 3 Llama 3.2 3B

3.2B

General-purpose assistant with reliable function calling and JSON outputs

131K contextiPhone & iPad · Mac

[View details](/models/hermes-3-llama-3.2-3b)

![Meta logo](/model-logos/meta.svg)

### Meta Llama 3.2 3B Instruct

3.2B

General-purpose model in the Llama 3.2 3B family

iPhone & iPad · Mac

[View details](/models/llama-3.2-3b-instruct)

![Qwen logo](/model-logos/qwen.svg)

### Dolphin 3.0 Qwen 2.5 3B

3.1B

Uncensored Qwen 2.5 model with full system prompt steerability

33K contextiPhone & iPad · Mac

[View details](/models/dolphin3.0-qwen2.5-3b)

![Qwen logo](/model-logos/qwen.svg)

### Qwen 2.5 3B

3.1B

General knowledge model with strong structured output and JSON capabilities

33K contextiPhone & iPad · Mac

[View details](/models/qwen2.5-3b-instruct)

![Qwen logo](/model-logos/qwen.svg)

### Qwen 2.5 Coder 3B

3.1B

Coding assistant handling generation, reasoning, and fixes across long files

33K contextiPhone & iPad · Mac

[View details](/models/qwen2.5-coder-3b-instruct)

![Meta logo](/model-logos/meta.svg)

### Dolphin 3.0 Llama 3.2 3B

3B

Uncensored Llama 3.2 3B instruct with user-defined alignment and 131K context

131K contextiPhone & iPad · Mac

[View details](/models/dolphin3.0-llama3.2-3b)

![Meta logo](/model-logos/meta.svg)

### Llama 3.2 3B Instruct Uncensored

3B

Uncensored Llama 3.2 3B modification that bypasses refusal mechanisms

131K contextiPhone & iPad · Mac

[View details](/models/llama-3.2-3b-instruct-uncensored)

![Stability AI logo](/model-logos/stability.svg)

### Nous Capybara 3B V1.9

2.8B

Complex summaries of advanced topics from an instruction-tuned StableLM 3B assistant

4K contextiPhone & iPad

[View details](/models/nous-capybara-3b-v1.9)

![Stability AI logo](/model-logos/stability.svg)

### Rocket 3B

2.8B

Detailed general assistant finetuned from StableLM 3B with DPO

4K contextiPhone & iPad

[View details](/models/rocket-3b)

![Microsoft logo](/model-logos/microsoft.svg)

### Dolphin 2.6 Phi 2

2.8B

Uncensored conversation, roleplay, and agent tasks from a Phi 2 base

iPhone & iPad · Mac

[View details](/models/dolphin-2-6-phi-2)

![Microsoft logo](/model-logos/microsoft.svg)

### Phi 2 Orange

2.8B

General reasoning specialist finetune of Phi 2 3B

iPhone & iPad · Mac

[View details](/models/phi-2-orange)

![Microsoft logo](/model-logos/microsoft.svg)

### Phi 2 Orange v2

2.8B

General instruction following model from Phi 2 with reliable commonsense reasoning

2K contextiPhone & iPad · Mac

[View details](/models/phi-2-orange-v2)

![Microsoft logo](/model-logos/microsoft.svg)

### Phi 2 Super

2.8B

Instruction-following conversational finetune of Phi 2 3B with simple chat template

2K contextiPhone & iPad

[View details](/models/phi-2-super)

![Google Gemma logo](/model-logos/gemma.svg)

### Gemma 2 2B IT

2.6B

General-purpose model in the Gemma 2 2B family

iPhone & iPad · Mac

[View details](/models/gemma-2-2b-it)

![Google Gemma logo](/model-logos/gemma.svg)

### SauerkrautLM Gemma 2 2B IT

2.6B

General-purpose fine-tune of Gemma 2 2B with multilingual strengths

8K contextiPhone & iPad · Mac

[View details](/models/sauerkrautlm-gemma-2-2b-it)

![Google Gemma logo](/model-logos/gemma.svg)

### Gemma 1.1 2B IT

2.5B

General-purpose model in the Gemma family

iPhone & iPad · Mac

[View details](/models/gemma-1.1-2b-it)

![Google Gemma logo](/model-logos/gemma.svg)

### Gemma 2B IT

2.5B

General-purpose model in the Gemma family

iPhone & iPad · Mac

[View details](/models/gemma-2b-it)

### H2O Danube 1.8B Chat

1.8B

General chat assistant fine-tuned for instruction following and factual accuracy

16K contextiPhone & iPad

[View details](/models/h2o-danube-1.8b-chat)

![Qwen logo](/model-logos/qwen.svg)

### EVA D Qwen2.5 1.5B v0.0

1.8B

Roleplay-focused distillation of Qwen 2.5 with 131K context

131K contextiPhone & iPad · Mac

[View details](/models/eva-d-qwen2.5-1.5b-v0.0)

![Stability AI logo](/model-logos/stability.svg)

### StableLM 2 Zephyr 1.6B

1.6B

General instruction assistant from StableLM 3B outperforming 10x larger models

4K contextiPhone & iPad

[View details](/models/stablelm-2-zephyr-1-6b)

![Qwen logo](/model-logos/qwen.svg)

### Dolphin 3.0 Qwen 2.5 1.5B

1.5B

Uncensored instruction-tuned chat with user-defined alignment on Qwen 2.5 1.5B

131K contextiPhone & iPad · Mac

[View details](/models/dolphin3.0-qwen2.5-1.5b)

![Qwen logo](/model-logos/qwen.svg)

### Qwen 2.5 1.5B

1.5B

Instruction-tuned general assistant reliably producing clean JSON

33K contextiPhone & iPad · Mac

[View details](/models/qwen2.5-1.5b-instruct)

![Qwen logo](/model-logos/qwen.svg)

### Qwen 2.5 Coder 1.5B

1.5B

Code completion, refactoring, and explanation expert from Qwen 2.5

33K contextiPhone & iPad · Mac

[View details](/models/qwen2.5-coder-1.5b-instruct)

![Meta logo](/model-logos/meta.svg)

### Llama 3.2 1B Instruct Abliterated

1.5B

Uncensored model derived by abliterating Llama 3.2 1B Instruct

131K contextiPhone & iPad · Mac

[View details](/models/llama-3.2-1b-instruct-abliterated)

![Meta logo](/model-logos/meta.svg)

### FuseChat Llama 3.2 1B Instruct

1.2B

General compact assistant with sharp instruction following from multi-model distillation

131K contextiPhone & iPad · Mac

[View details](/models/fusechat-llama-3.2-1b-instruct)

![Meta logo](/model-logos/meta.svg)

### Meta Llama 3.2 1B Instruct

1.2B

General-purpose model in the Llama 3.2 1B family

iPhone & iPad · Mac

[View details](/models/llama-3.2-1b-instruct)

### TinyDolphin 2.8 1.1B Chat

1.1B

Uncensored conversation specialist built on TinyLlama 1.1B

4K contextiPhone & iPad

[View details](/models/tinydolphin-2.8-1.1b)

### TinyLlama 1.1B Chat

1.1B

Conversational assistant fine-tuned from TinyLlama 1B for on-device use

2K contextiPhone & iPad

[View details](/models/tinyllama-1.1b-chat-v1.0)

![Meta logo](/model-logos/meta.svg)

### Dolphin 3.0 Llama 3.2 1B

1B

Uncensored instruct finetune of Llama 3.2 1B for coding and agentic tasks

131K contextiPhone & iPad · Mac

[View details](/models/dolphin3.0-llama3.2-1b)

![Google Gemma logo](/model-logos/gemma.svg)

### Amoral Gemma 3 1B v2

1B

Uncensored Gemma 3 1B finetune for neutral factual responses

33K contextiPhone & iPad · Mac

[View details](/models/amoral-gemma3-1b-v2)

![Google Gemma logo](/model-logos/gemma.svg)

### Gemma 3 1B IT

1B

General-purpose model in the Gemma 3 1B family

iPhone & iPad · Mac

[View details](/models/gemma-3-1b-it)

![Google Gemma logo](/model-logos/gemma.svg)

### Gemma 3 1B IT Abliterated

1B

Uncensored Gemma 3 1B via abliteration

33K contextiPhone & iPad · Mac

[View details](/models/gemma-3-1b-it-abliterated)

![Qwen logo](/model-logos/qwen.svg)

### Dolphin 3.0 Qwen 2.5 0.5B

0.5B

Uncensored Qwen 2.5 0.5B instruct with full system prompt control

33K contextiPhone & iPad · Mac

[View details](/models/dolphin3.0-qwen2.5-0.5b)

![Qwen logo](/model-logos/qwen.svg)

### Qwen 2.5 0.5B Unquantized

0.5B

General-purpose 0.5B model with strong instruction following and JSON output

33K contextiPhone & iPad · Mac

[View details](/models/qwen2.5-0.5b-instruct)

![Qwen logo](/model-logos/qwen.svg)

### Qwen 2.5 Coder 0.5B Unquantized

0.5B

Code generation, reasoning, and repair via Qwen 2.5 with 32K context

33K contextiPhone & iPad · Mac

[View details](/models/qwen2.5-coder-0.5b-instruct)

[Refine in the full gallery](/models#models-gallery)

## Frequently asked questions

-   ### What is the largest model my iPhone can run?
    
    On iPhone, Private LLM runs models up to Hermes 2 Pro Llama 3 8B, depending on how much memory your specific iPhone has.
    
-   ### How many models can run on iPhone?
    
    Private LLM runs 88 models on iPhone, all fully on-device.
    
-   ### Do these models run offline on iPhone?
    
    Yes. Once downloaded, every model runs 100% on-device on your iPhone — no internet connection and no data sharing.