Private LLM (@private

Private LLM

541 posts

Private LLM

@private_llm

Local AI for Private, Uncensored Chat on iPhone, iPad, and Mac. No Cloud, No Tracking, No Logins. apple.co/4kqGUzx

Dublin, Ireland

Joined April 2023

Pinned
Private LLM
@private_llm
Jan 23, 2025
Replying to @private_llm
Can an iPhone hold its own against a Mac? 💻📱 Ollama is running DeepSeek R1 Distill Llama 8B on a 16GB M2 MacBook Air, while the iPhone 15 Pro (8GB A17 Pro) runs the same model. Same model. Same sampling temperature. Same system prompt. While the iPhone is a bit slower, the
00:00
26K
Private LLM
@private_llm
Jun 2, 2023
Look ma, no Internet!
00:00
45K
Private LLM
@private_llm
Dec 11, 2024
Way more than yours x.com/localghost/sta…
00:00
3.3K
Private LLM
@private_llm
May 3, 2024
🤣
102K
Private LLM
@private_llm
Jan 22, 2025
⚡️New "Thinking" UI with DeepSeek R1 Distilled models, running completely locally on iPhone
00:00
189K
Private LLM
@private_llm
Dec 10, 2024
The difference is quite simple. Apollo is a run of the mill MLX wrapper with two tiny RTN quantized Llama 3.2 based models, while Private LLM for iOS has 64 downloadable OmniQuant and GPTQ quantized models. The macOS version of Private LLM has many more, but it wouldn’t be fair
00:00
Brandon McNew - Domesticaveman
@domesticaveman
Dec 9, 2024
Replying to @localghost and @apolloaiapp
I'm wondering how @apolloaiapp differs from @private_llm. Do they function similarly? i'm definitely interested in yours, and this concept in general. Just used ollama to download llama3.3 and that LLM was a massive RAM liability even with 64GB lol
4.2K
Private LLM
@private_llm
Dec 24, 2024
Llama 3.3 can answer this fully offline, in a few seconds, without a $200/month subscription and over a minute of “thinking”. 😂
Kay
@kaysorin
Dec 23, 2024
last night i demo’d @openai o1 pro to my most skeptical friend - the results were extraordinary. it was amazing to see o1 pro successfully tackle humanities research, and how much better it was than 4o / o1. i also had a ton of fun watching my friend come around to AI! story ⬇️
16K
Private LLM
@private_llm
Apr 20, 2024
Private LLM for iOS v1.7.8 is now live on the App Store. 🎉 Experience the power of the latest Llama 3 8B Instruct model from @AIatMeta, running privately, fully on-device with no internet connection or telemetry. Works on all Pro, Pro Max iPhones and Apple Silicon iPads. Also,
00:00
44K
Private LLM
@private_llm
Jan 5, 2024
Private LLM for macOS v1.6.2 was just released. This release adds a 4-bit OmniQuant quantized version of the incredible new Nous-Hermes-2-SOLAR-10.7B model by @teknium. Full 4k context length with a memory footprint of only 6.85GB! 🎉
10K
Private LLM
@private_llm
Jan 23, 2025
We tested DeepSeek R1 Distill Llama models (8B & 70B) on math, reasoning, and censorship. But can an iPhone really reason as well as Ollama and LM Studio on a PC? The results will shock you. Thread🧵
20K
Private LLM
@private_llm
Apr 24, 2024
We’ve never seen a 7B model get the famous Sally test right, let alone a 3.8B model. 💪Even the Llama 3 8B Instruct model fails this test. Coming 🔜 to your 📱
00:00
Microsoft
@Microsoft
Apr 23, 2024
We're excited to announce the launch of Phi-3, a groundbreaking family of small language models that outperform larger models on a range of benchmarks. Learn how these small language models trained on high-quality data are doing more with less: msft.it/6010YHP32
37K
Private LLM
@private_llm
Apr 14, 2024
Private LLM v1.8.4 for macOS is here with three new models: - Gemma 1.1 2B IT from @GoogleDeepMind available on all compatible Intel and Apple Silicon Macs. - Dolphin 2.6 Mixtral 8x7B by @erhartford on Apple Silicon Macs with 32GB+ RAM. - Nous Hermes 2 Mixtral 8x7B DPO also
00:00
17K
Private LLM
@private_llm
Jan 24, 2025
This is why we're shipping uncensored DeepSeek R1 Distill models soon! Drop a comment with prompts you'd like to see tested 👇
00:00
00:05
Jane Manchun Wong
@wongmjane
Jan 24, 2025
DeepSeek censors its own response in realtime as soon as Xi Jinping is mentioned
32K
Private LLM
@private_llm
May 9, 2024
🖥️ v1.8.6 of Private LLM for macOS was released! 🎉 This release adds a 4-bit OmniQuant quantized version of the SOTA Llama 3 Instruct 70B🦙model by @AIatMeta. Works on all Apple Silicon Macs with 48GB or more RAM.
27K