Video yükleniyor...

Video Yüklenemedi

Ana Sayfaya Dön

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 📱

private_llm's profile picture

Private LLM

3,910 subscribers

36,953 görüntüleme • 2 yıl önce •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

8 Yorum

Private LLM profil fotoğrafı
Private LLM2 yıl önce

It's now live on the App Store. 🚀

ANDREW ST JOHN profil fotoğrafı
ANDREW ST JOHN2 yıl önce

Can’t wait to try it! How are the hallucinations?

Private LLM profil fotoğrafı
Private LLM2 yıl önce

You can download it from the App Store now

Mattias Lamotte profil fotoğrafı
Mattias Lamotte2 yıl önce

Looking forward to testing it out! How long will it take before it appears in the app? I guess you have to upgrade the app in the app store which takes a couple of weeks to get approved?

Private LLM profil fotoğrafı
Private LLM2 yıl önce

We've submitted the app for review and it should either today or tomorrow.

Private LLM profil fotoğrafı
Private LLM2 yıl önce

That’s a great question! It’s a part of an ad-hoc suite benchmarks prompts that are usually run to evaluate LLMs. Until recently, the smallest LLM that could get it right was the much bigger Mixtral-8x7B model, which we support on the macOS version of Private LLM, but it’s too big to fit on iPhones. This model changes that!

George P. profil fotoğrafı
George P.2 yıl önce

is there an Android app in the works?

Private LLM profil fotoğrafı
Private LLM2 yıl önce

@Angel_pap85 I wish there was one. No, not yet.

Benzer Videolar

Introducing the OpenRouter MCP, live model intelligence right inside your agent Your agent builds and ships, but when it comes to choosing the right model for the right job, it guesses from 6 month old training data Watch it pick, price, and test the right model:
2:30

Introducing the OpenRouter MCP, live model intelligence right inside your agent Your agent builds and ships, but when it comes to choosing the right model for the right job, it guesses from 6 month old training data Watch it pick, price, and test the right model:

OpenRouter

264,251 görüntüleme • 1 ay önce

This is a slalom test image of the Model YL. It passed the test at 110 km/h.
0:28

This is a slalom test image of the Model YL. It passed the test at 110 km/h.

Tsla Chan

275,437 görüntüleme • 1 yıl önce

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 3-bit OmniQuant model reasons just as well as the 4-bit RTN model on the Mac. 🤯
1:07

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 3-bit OmniQuant model reasons just as well as the 4-bit RTN model on the Mac. 🤯

Private LLM

26,544 görüntüleme • 1 yıl önce

Sam Altman's test for whether your startup survives the next AI model: "Ask if a 100x model improvement excites them, or scares them"
2:28

Sam Altman's test for whether your startup survives the next AI model: "Ask if a 100x model improvement excites them, or scares them"

Big Brain AI

22,672 görüntüleme • 4 ay önce

End to End Speech models are on fire - LLAMA-OMNI 8B - Apache licensed! 🔥 > Speech Encoder - Whisper Large v3 > LLM backbone - Llama 3.1 8B Instruct > Speech Decoder - HuBERT (UnitY) > Simultaneously generate Speech + Text > Less than 250 ms latency > Trained in less than 3 days on 4x GPUs > Used 200K instruct pairs > Model checkpoints on the Hub 🤗 > Space incoming! GG! I'm here for this trend! 🐐
0:36

End to End Speech models are on fire - LLAMA-OMNI 8B - Apache licensed! 🔥 > Speech Encoder - Whisper Large v3 > LLM backbone - Llama 3.1 8B Instruct > Speech Decoder - HuBERT (UnitY) > Simultaneously generate Speech + Text > Less than 250 ms latency > Trained in less than 3 days on 4x GPUs > Used 200K instruct pairs > Model checkpoints on the Hub 🤗 > Space incoming! GG! I'm here for this trend! 🐐

Vaibhav (VB) Srivastav

47,921 görüntüleme • 1 yıl önce

The Tesla Model Y “Packs” a punch when it comes to space. No other SUV in this category comes close to this, let alone even bigger midsize SUV’s. There’s so much to love about the Tesla Model Y!
0:39

The Tesla Model Y “Packs” a punch when it comes to space. No other SUV in this category comes close to this, let alone even bigger midsize SUV’s. There’s so much to love about the Tesla Model Y!

Bearded Tesla

114,208 görüntüleme • 7 ay önce

MMD worlds are fun! (Great way to test a model too!) This bee is still WIP but it's coming along!
0:45

MMD worlds are fun! (Great way to test a model too!) This bee is still WIP but it's coming along!

Modern

20,390 görüntüleme • 3 yıl önce

Last before bed - A little action sequence I wanted to test out using the new Flux 3 Black Forest Labs. This model has range.
0:20

Last before bed - A little action sequence I wanted to test out using the new Flux 3 Black Forest Labs. This model has range.

Sam Wasserman

12,034 görüntüleme • 26 gün önce

Dubai was long seen as a safe haven in an uncertain part of the world. But the Iran war has put that model to the test
1:40

Dubai was long seen as a safe haven in an uncertain part of the world. But the Iran war has put that model to the test

Bloomberg

85,011 görüntüleme • 5 ay önce

Introducing Merge Gateway - Build Your Own Router. You're three sprints into your coding assistant. You pick the most hyped model, integrate, test, deploy. A month later, a new model drops. Now you re-test, re-integrate, re-deploy. Your product didn't change, but the benchmark did. That's how most AI teams operate. Chasing a "best" defined by people who've never seen their product. There is no best model. There's only the right one for your product, users, and use-cases. Build Your Own Router runs on your definition of good. Pick your benchmarks, weigh them, add your own evals. Merge routes every request to your winner. 👉$100 in credits to the first 200 people that comment
1:57

Introducing Merge Gateway - Build Your Own Router. You're three sprints into your coding assistant. You pick the most hyped model, integrate, test, deploy. A month later, a new model drops. Now you re-test, re-integrate, re-deploy. Your product didn't change, but the benchmark did. That's how most AI teams operate. Chasing a "best" defined by people who've never seen their product. There is no best model. There's only the right one for your product, users, and use-cases. Build Your Own Router runs on your definition of good. Pick your benchmarks, weigh them, add your own evals. Merge routes every request to your winner. 👉$100 in credits to the first 200 people that comment

Shensi Ding

3,666,966 görüntüleme • 2 ay önce

A lot of people reporting that shields colliders don't represent what is seen on the model are a correct, quick test with bots. #Rust
1:29

A lot of people reporting that shields colliders don't represent what is seen on the model are a correct, quick test with bots. #Rust

Protox

35,828 görüntüleme • 5 ay önce

$TSLA Model 3 is the only vehicle that has passed a drill-and-crash truck test, and it's getting a lot of safety attention in China. Watch the new video I just uploaded, and compare the vehicles. Other cars aren't even protected by the A-pillar in the first place. We have to choose the safest vehicle for our family. (Model Y does not participate in the test)
2:45

$TSLA Model 3 is the only vehicle that has passed a drill-and-crash truck test, and it's getting a lot of safety attention in China. Watch the new video I just uploaded, and compare the vehicles. Other cars aren't even protected by the A-pillar in the first place. We have to choose the safest vehicle for our family. (Model Y does not participate in the test)

Tsla Chan

3,286,647 görüntüleme • 1 yıl önce