Загрузка видео...

Не удалось загрузить видео

На главную

Running Gemma 4 12B on your iPhone? Yes! 🧵 LM Studio + Locally AI latest version with LM Link is really cool! This opens up additional scenarios! My brain is on fire 🔥

ivanfioravanti's profile picture

Ivan Fioravanti ᯅ

38,506 subscribers

19,292 просмотров • 2 месяцев назад •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 0

Нет доступных комментариев

Здесь появятся комментарии из оригинального поста

Похожие видео

Testing the new Gemma 4 12B (QAT) vision and OCR capabilities locally with LM Studio. # The setup: - GPU: NVIDIA RTX 4060 (8GB VRAM) - CPU: Intel i7 - Runner: LM Studio - Config: 32k context, 38 layers offloaded, Flash Attention enabled - Speed: ~14 tokens/sec decode throughput # The test: I gave it a screenshot of Google AI Studio. Prompt: "clone this. give me a single html file" # The result: A solid one shot replication. It successfully mapped out the layout, recognized the UI text, and structured the divs correctly, with only minor differences from the original. Results available at the end of the video. Quite capable for a 12B model running on budget consumer hardware. A gpu that costs only $300. # Why the architecture under the hood is notable: Unlike traditional models that rely on heavy, separate vision and audio encoders, Gemma 4 12B uses a unified, encoder free architecture. It bypasses separate multi stage encoders. Uses a 35M parameter vision embedder to project raw 48x48 pixel patches directly to the LLM hidden dimension. Local multimodal development is becoming highly accessible on standard hardware. If you've spun up Gemma 4 12B locally, what setup are you using and what kind of throughput are you seeing?
0:16

Testing the new Gemma 4 12B (QAT) vision and OCR capabilities locally with LM Studio. # The setup: - GPU: NVIDIA RTX 4060 (8GB VRAM) - CPU: Intel i7 - Runner: LM Studio - Config: 32k context, 38 layers offloaded, Flash Attention enabled - Speed: ~14 tokens/sec decode throughput # The test: I gave it a screenshot of Google AI Studio. Prompt: "clone this. give me a single html file" # The result: A solid one shot replication. It successfully mapped out the layout, recognized the UI text, and structured the divs correctly, with only minor differences from the original. Results available at the end of the video. Quite capable for a 12B model running on budget consumer hardware. A gpu that costs only $300. # Why the architecture under the hood is notable: Unlike traditional models that rely on heavy, separate vision and audio encoders, Gemma 4 12B uses a unified, encoder free architecture. It bypasses separate multi stage encoders. Uses a 35M parameter vision embedder to project raw 48x48 pixel patches directly to the LLM hidden dimension. Local multimodal development is becoming highly accessible on standard hardware. If you've spun up Gemma 4 12B locally, what setup are you using and what kind of throughput are you seeing?

Alok

25,717 просмотров • 2 месяцев назад

It’s never been easier to have your own private AIs running at home, off-grid, or anywhere else. For AI independence, all you need is: - LM Studio on your computer. - Apollo in your pocket.
0:13

It’s never been easier to have your own private AIs running at home, off-grid, or anywhere else. For AI independence, all you need is: - LM Studio on your computer. - Apollo in your pocket.

Aaron Ng

13,319 просмотров • 1 год назад

Want to teach Gemma to master chess? Check out this awesome community project showing how to fine-tune Gemma 4 12B on your own data, 100% locally! Running text, images, and audio on just 8GB VRAM makes custom models more accessible than ever.
0:11

Want to teach Gemma to master chess? Check out this awesome community project showing how to fine-tune Gemma 4 12B on your own data, 100% locally! Running text, images, and audio on just 8GB VRAM makes custom models more accessible than ever.

Google Gemma

157,609 просмотров • 2 месяцев назад

We made it easy to run local models on your computer and use them from your phone. A secure, end-to-end encrypted connection in just a few clicks. Available now with LM Studio and Locally AI - Local AI Chat.
0:16

We made it easy to run local models on your computer and use them from your phone. A secure, end-to-end encrypted connection in just a few clicks. Available now with LM Studio and Locally AI - Local AI Chat.

Adrien Grondin

12,045 просмотров • 2 месяцев назад

Google’s Gemma 4 E2B running on-device on iPhone 17 Pro Gemma 4 is built from the same research as Gemini 3, has image understanding capabilities and can reason if needed Running at ~40tk/s with MLX optimized for Apple Silicon
0:12

Google’s Gemma 4 E2B running on-device on iPhone 17 Pro Gemma 4 is built from the same research as Gemini 3, has image understanding capabilities and can reason if needed Running at ~40tk/s with MLX optimized for Apple Silicon

Adrien Grondin

1,044,994 просмотров • 4 месяцев назад

One of the coolest uses of LM Studio and AI at AMD is using LM Link to have a Framework Desktop be a remote inference server for a laptop, including something little like Framework Laptop 12. It feels native on the laptop, but gets the desktop’s performance.
0:18

One of the coolest uses of LM Studio and AI at AMD is using LM Link to have a Framework Desktop be a remote inference server for a laptop, including something little like Framework Laptop 12. It feels native on the laptop, but gets the desktop’s performance.

Framework

39,079 просмотров • 5 месяцев назад

$AMD just put out this post saying: “Deploying Deepseek R1 distilled "reasoning" models on AMD Ryzen AI processors and Radeon graphics cards is incredibly easy and available now through LM Studio”
0:15

$AMD just put out this post saying: “Deploying Deepseek R1 distilled "reasoning" models on AMD Ryzen AI processors and Radeon graphics cards is incredibly easy and available now through LM Studio”

Evan

171,522 просмотров • 1 год назад

Apple wasn’t kidding, the iPhone 17 Pro is really built for running LLMs Here’s LFM2 8B A1B by running on-device with MLX in Locally AI - Local AI Chat, the iPhone runs the 8B model with zero struggle Thanks Prince Canuma for the port to MLX, it made the MLX Swift port possible
0:17

Apple wasn’t kidding, the iPhone 17 Pro is really built for running LLMs Here’s LFM2 8B A1B by running on-device with MLX in Locally AI - Local AI Chat, the iPhone runs the 8B model with zero struggle Thanks Prince Canuma for the port to MLX, it made the MLX Swift port possible

Adrien Grondin

493,561 просмотров • 10 месяцев назад

i just ran Google's brand new Unsloth Gemma4 12B dense GGUF on my RTX 4060 using llama.cpp + CUDA 13.2 21 tokens per second. on a budget consumer GPU. locally. no API. no cloud. no subscription. and the benchmarks are absolutely cooked # first let's talk architecture because this is genuinely different every multimodal model you've used has a frozen vision encoder + frozen audio encoder + LLM backbone glued together Gemma 4 12B is different it's a single decoder only transformer. that's it. vision? raw 48×48 pixel patches → one matmul → projected directly into the LLM audio? raw 16kHz signal sliced into 40ms frames → linear projection → same LLM input space no encoder tax. no latency penalty. no fragmented memory to put the encoder savings in perspective: old Gemma 4 26B approach: - 550M param vision encoder (frozen) - 300M param audio encoder (frozen) - LLM backbone Gemma 4 12B: - 35M param vision embedder (a single matmul) - no audio encoder at all - LLM backbone handles EVERYTHING 550M → 35M for vision alone. that's a 15x reduction this is why the gemma-4-12b-it-Q4_K_M.gguf is just 6.6 GBs!!! and it has 256K native context context # Benchmarks: AIME 2026 (math olympiad): 77.5% GPQA Diamond (expert science): 78.8% LiveCodeBench v6 (real code): 72% Codeforces ELO: 1659 MMLU Pro: 77.2% MATH-Vision: 79.7% BigBench Extra Hard: 53% inference → llama.cpp, LM Studio, vLLM, SGLang llamacpp flags: -m "gemma-4-12b-it-Q4_K_M.gguf" -ngl 99 -c 8000 -v --port 8080 Available on huggingface now! Link below
0:18

i just ran Google's brand new Unsloth Gemma4 12B dense GGUF on my RTX 4060 using llama.cpp + CUDA 13.2 21 tokens per second. on a budget consumer GPU. locally. no API. no cloud. no subscription. and the benchmarks are absolutely cooked # first let's talk architecture because this is genuinely different every multimodal model you've used has a frozen vision encoder + frozen audio encoder + LLM backbone glued together Gemma 4 12B is different it's a single decoder only transformer. that's it. vision? raw 48×48 pixel patches → one matmul → projected directly into the LLM audio? raw 16kHz signal sliced into 40ms frames → linear projection → same LLM input space no encoder tax. no latency penalty. no fragmented memory to put the encoder savings in perspective: old Gemma 4 26B approach: - 550M param vision encoder (frozen) - 300M param audio encoder (frozen) - LLM backbone Gemma 4 12B: - 35M param vision embedder (a single matmul) - no audio encoder at all - LLM backbone handles EVERYTHING 550M → 35M for vision alone. that's a 15x reduction this is why the gemma-4-12b-it-Q4_K_M.gguf is just 6.6 GBs!!! and it has 256K native context context # Benchmarks: AIME 2026 (math olympiad): 77.5% GPQA Diamond (expert science): 78.8% LiveCodeBench v6 (real code): 72% Codeforces ELO: 1659 MMLU Pro: 77.2% MATH-Vision: 79.7% BigBench Extra Hard: 53% inference → llama.cpp, LM Studio, vLLM, SGLang llamacpp flags: -m "gemma-4-12b-it-Q4_K_M.gguf" -ngl 99 -c 8000 -v --port 8080 Available on huggingface now! Link below

Alok

281,007 просмотров • 2 месяцев назад

Claude fable 5 is cooked Claude fable 5 Vs Gemma 4 26b a4b qat MTP Gemma 4 26b running locally on my 8GB vram single RTX 4060 built this using three.js in a single session and 3 prompts. no cloud, no subscription 100% private unlimited use. how long until it can catch up completely?
0:19

Claude fable 5 is cooked Claude fable 5 Vs Gemma 4 26b a4b qat MTP Gemma 4 26b running locally on my 8GB vram single RTX 4060 built this using three.js in a single session and 3 prompts. no cloud, no subscription 100% private unlimited use. how long until it can catch up completely?

Alok

30,802 просмотров • 2 месяцев назад

Tried Gemma 4 ran locally on my iPhone today I thought it'd be useful in case the apocalypse happens and I need to ask it for survival tips Like how to make a fire 🔥 I guess I'll freeze to death instead 🫠
0:10

Tried Gemma 4 ran locally on my iPhone today I thought it'd be useful in case the apocalypse happens and I need to ask it for survival tips Like how to make a fire 🔥 I guess I'll freeze to death instead 🫠

@levelsio

614,706 просмотров • 4 месяцев назад

This is the most hilarious thing I saw and did today Ran gemma-4-12B-coder-fable5-composer2.5-v1-GGUF locally with 8 GB VRAM at 20+ tok/sec Anthropic's Claude Fable 5 launched June 9. By June 12 it was banned. I can't access it. You can't either. But here's the twist: I'm running a model trained on its chain of thought at 20 tok/s on my RTX 4060 8GB. Locally. Offline. No cloud. No export control. Enter: Gemma4-12B-Coder GGUF (Q4_K_M) Base: Google's gemma-4-12B-it Fine-tuned on verifiable Python CoT data: - Primary: Composer 2.5 real reasoning traces (only passing solutions kept) - Auxiliary: Fable 5 used to redo the hard cases Composer missed. Every training example's reasoning led to code that actually ran. No hallucinated logic. Llama.cpp flags: -m gemma4-coding-Q4_K_M.gguf -cnv -ngl 44 -c 64000 -v (huggingface model link in comments) Flag breakdown: -ngl 44 → offload 44 layers to GPU (tune this for your VRAM) -c 64000 → 64K context window -cnv → conversation/chat mode -v → verbose output The irony writes itself. Anthropic spent weeks telling the world Fable 5 (mythos) is too powerful to release. Then released it. Then got banned from serving it, including their own researchers. Meanwhile: a Gemma 4 12B fine tune, trained on Fable 5's reasoning, runs fully offline on my mid range consumer GPU No API. No cloud. Just me and llama.cpp. This is why local AI matters. Check out the model's link in the comments. How's your experience been with this model?
0:15

This is the most hilarious thing I saw and did today Ran gemma-4-12B-coder-fable5-composer2.5-v1-GGUF locally with 8 GB VRAM at 20+ tok/sec Anthropic's Claude Fable 5 launched June 9. By June 12 it was banned. I can't access it. You can't either. But here's the twist: I'm running a model trained on its chain of thought at 20 tok/s on my RTX 4060 8GB. Locally. Offline. No cloud. No export control. Enter: Gemma4-12B-Coder GGUF (Q4_K_M) Base: Google's gemma-4-12B-it Fine-tuned on verifiable Python CoT data: - Primary: Composer 2.5 real reasoning traces (only passing solutions kept) - Auxiliary: Fable 5 used to redo the hard cases Composer missed. Every training example's reasoning led to code that actually ran. No hallucinated logic. Llama.cpp flags: -m gemma4-coding-Q4_K_M.gguf -cnv -ngl 44 -c 64000 -v (huggingface model link in comments) Flag breakdown: -ngl 44 → offload 44 layers to GPU (tune this for your VRAM) -c 64000 → 64K context window -cnv → conversation/chat mode -v → verbose output The irony writes itself. Anthropic spent weeks telling the world Fable 5 (mythos) is too powerful to release. Then released it. Then got banned from serving it, including their own researchers. Meanwhile: a Gemma 4 12B fine tune, trained on Fable 5's reasoning, runs fully offline on my mid range consumer GPU No API. No cloud. Just me and llama.cpp. This is why local AI matters. Check out the model's link in the comments. How's your experience been with this model?

Alok

572,538 просмотров • 2 месяцев назад

Batching for vision models is now available in Beta with our latest MLX engine update 👾 The updated engine also brings major improvements to caching for faster inference overall. Turn on Developer Mode, choose the beta runtime channel, and select LM Studio MLX v1.8.1.
0:11

Batching for vision models is now available in Beta with our latest MLX engine update 👾 The updated engine also brings major improvements to caching for faster inference overall. Turn on Developer Mode, choose the beta runtime channel, and select LM Studio MLX v1.8.1.

LM Studio

47,996 просмотров • 3 месяцев назад

Hope is keeping it cool while everything’s on fire with the classic words, ‘This is fine, everything is fine.’ 🔥If you vibe with that energy, grab your ‘This is Fine’ merch from our official site:
0:18

Hope is keeping it cool while everything’s on fire with the classic words, ‘This is fine, everything is fine.’ 🔥If you vibe with that energy, grab your ‘This is Fine’ merch from our official site:

Max Design Pro

64,540 просмотров • 1 год назад

Gemma 4 analyzes the video. Generates key questions. Calls Falcon Perception. "Find all the people." 156 found. "Detect only white cars." 8 found. A 26B model is running agentic multi-QA vision orchestration. The models are running locally on a MacBook with MLX. No API.
0:16

Gemma 4 analyzes the video. Generates key questions. Calls Falcon Perception. "Find all the people." 156 found. "Detect only white cars." 8 found. A 26B model is running agentic multi-QA vision orchestration. The models are running locally on a MacBook with MLX. No API.

Maziyar PANAHI

158,988 просмотров • 4 месяцев назад

laziest meme evr this is ai generated btw #coinpin #bfdia yes i made this on my phone shut up
0:13

laziest meme evr this is ai generated btw #coinpin #bfdia yes i made this on my phone shut up

ayden

23,270 просмотров • 7 месяцев назад

GEMMA 4 26B ON AN RTX 4060 WITH A 248K TOKEN CONTEXT WINDOW 20 tokens per second and a context window so large you can feed it entire codebases, books and research papers in a single prompt this is not a cloud api and not a server rack, this is a regular consumer gpu running locally with llama.cpp and q4_k_xl quantization 248k context on an 8gb vram card was not supposed to be possible and here it is just running on someone’s desk the article below covers exactly which tools and configs make this kind of setup work in 2026 ↓
0:19

GEMMA 4 26B ON AN RTX 4060 WITH A 248K TOKEN CONTEXT WINDOW 20 tokens per second and a context window so large you can feed it entire codebases, books and research papers in a single prompt this is not a cloud api and not a server rack, this is a regular consumer gpu running locally with llama.cpp and q4_k_xl quantization 248k context on an 8gb vram card was not supposed to be possible and here it is just running on someone’s desk the article below covers exactly which tools and configs make this kind of setup work in 2026 ↓

leopardracer

56,772 просмотров • 2 месяцев назад