Loading video...

Video Failed to Load

Go Home

MAI-Code-1.1-Flash is going local with on-device model calls inside Github Copilot. High-quality agentic coding on your device, with no inference charge for local model calls.

MicrosoftAI's profile picture

Microsoft AI

21,869 subscribers

126,275 views • 6 days ago •via X (Twitter)

Science & Technology
LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

First steps for a specialized DeepSeek v4 Flash inference engine focused on inference quality / stability at different quantizations, with networked API that is batching capable. This is the 2 bit quants model running on my M3 Max 128GB.
0:12

First steps for a specialized DeepSeek v4 Flash inference engine focused on inference quality / stability at different quantizations, with networked API that is batching capable. This is the 2 bit quants model running on my M3 Max 128GB.

antirez

14,220 views • 5 months ago

QVAC Health new beta supporting local on device model to analyze food images, calculate calories and nutrients etc 100% privacy. Getting ready for public beta.
0:17

QVAC Health new beta supporting local on device model to analyze food images, calculate calories and nutrients etc 100% privacy. Getting ready for public beta.

Paolo Ardoino 🤖

31,381 views • 1 year ago

Cool: Hugging Face is now supported in opencode, select your favorite model from the Hub and it will consume your inference quota directly. I recommend starting with GLM-4.5 (great at coding + awesome at agentic tasks)
0:14

Cool: Hugging Face is now supported in opencode, select your favorite model from the Hub and it will consume your inference quota directly. I recommend starting with GLM-4.5 (great at coding + awesome at agentic tasks)

Victor M

50,554 views • 1 year ago

Gemini Omni Flash 1.1 is now available on Google's preview model list A cheap AI video model that could directly compete with Seedance 2.5 But if its quality keeps getting better and cheaper, AI video will soon be VERY exciting. But the real question is: Do cheap, high-quality AI videos devalue human artists or empower them? I tried out Gemini Omni Flash 1.1. And I must say that I was shocked by the results
0:10

Gemini Omni Flash 1.1 is now available on Google's preview model list A cheap AI video model that could directly compete with Seedance 2.5 But if its quality keeps getting better and cheaper, AI video will soon be VERY exciting. But the real question is: Do cheap, high-quality AI videos devalue human artists or empower them? I tried out Gemini Omni Flash 1.1. And I must say that I was shocked by the results

Priy

53,006 views • 1 month ago

Woow Tencent has released a model 100% open source which is better than Google Translate... And it runs locally on your phone?! - Only 440MB (!!) - 33 languages - On-device translation So we have now a local Google Translate which is running completely offline. They are using AngelSlim which their compression toolkit. That's why the model is so efficient.
0:12

Woow Tencent has released a model 100% open source which is better than Google Translate... And it runs locally on your phone?! - Only 440MB (!!) - 33 languages - On-device translation So we have now a local Google Translate which is running completely offline. They are using AngelSlim which their compression toolkit. That's why the model is so efficient.

Paul Couvert

142,678 views • 5 months ago

QVAC is AI without the cloud—modular, local, unstoppable. No API keys. No gatekeepers. Just pure on device intelligence. What will you build with the QVAC SDK when it drops? 👇
0:11

QVAC is AI without the cloud—modular, local, unstoppable. No API keys. No gatekeepers. Just pure on device intelligence. What will you build with the QVAC SDK when it drops? 👇

QVAC

25,167 views • 1 year ago

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 views • 1 year ago

🤖 ExoBrain is a compute device built for large-scale AI collaboration, supporting model training, inference, and long-running Agents With a standardized architecture and continuous execution capability, it delivers efficient, reliable compute for AI systems ⚡ From one-time calls to continuous operation, ExoBrain enables AI to execute persistently, produce ongoing outputs, and generate long-term value #ExoBrain #AICompute #AIAgents #AGI
0:15

🤖 ExoBrain is a compute device built for large-scale AI collaboration, supporting model training, inference, and long-running Agents With a standardized architecture and continuous execution capability, it delivers efficient, reliable compute for AI systems ⚡ From one-time calls to continuous operation, ExoBrain enables AI to execute persistently, produce ongoing outputs, and generate long-term value #ExoBrain #AICompute #AIAgents #AGI

ExoBrain

21,550 views • 6 months ago

Routing for long-horizon coding agents is a big deal. Not Diamond just announced a model router that works natively with Claude Code. This is huge. It picks the model and reasoning effort before each turn in a session, runs through a privacy-preserving local proxy, and your requests still execute through your own gateway. In their benchmarks, it approximates Opus 4.8 Xhigh quality at 39 to 61 percent lower cost.
0:16

Routing for long-horizon coding agents is a big deal. Not Diamond just announced a model router that works natively with Claude Code. This is huge. It picks the model and reasoning effort before each turn in a session, runs through a privacy-preserving local proxy, and your requests still execute through your own gateway. In their benchmarks, it approximates Opus 4.8 Xhigh quality at 39 to 61 percent lower cost.

elvis

13,699 views • 2 months ago

i found a way to make UNCENSORED AI AGENT on a RTX 4090 GPU (!!!) with LOCAL 30B model weights this is GLM-4.7-Flash with abliteration, need 24GB VRAM, safety alignment surgically removed from the weights, the model has native tool calling, it actually executes bash, edits files, runs git (1) use ollama to pull weights of GLM > ollama pull huihui_ai/glm-4.7-flash-abliterated:q4_K (2) proxy it to any coding agent via ollama > ollama launch claude --model huihui_ai/glm-4.7-flash-abliterated:q4_K > ollama launch codex --model huihui_ai/glm-4.7-flash-abliterated:q4_K > ollama launch opencode --model huihui_ai/glm-4.7-flash-abliterated:q4_K (3) have fun
0:11

i found a way to make UNCENSORED AI AGENT on a RTX 4090 GPU (!!!) with LOCAL 30B model weights this is GLM-4.7-Flash with abliteration, need 24GB VRAM, safety alignment surgically removed from the weights, the model has native tool calling, it actually executes bash, edits files, runs git (1) use ollama to pull weights of GLM > ollama pull huihui_ai/glm-4.7-flash-abliterated:q4_K (2) proxy it to any coding agent via ollama > ollama launch claude --model huihui_ai/glm-4.7-flash-abliterated:q4_K > ollama launch codex --model huihui_ai/glm-4.7-flash-abliterated:q4_K > ollama launch opencode --model huihui_ai/glm-4.7-flash-abliterated:q4_K (3) have fun

chiefofautism

346,837 views • 7 months ago

🔥 GLM-5.3-Flash Hits #1 on 🐮 GLM-5.3-Flash (Ox Alpha) is now the most-used and most popular model on with cumulative token throughput surpassing 2.41 trillion tokens. As the first native omni-modal model in the GLM-5 series, it packs 320B total parameters, with 18B active, and features a hybrid sparse and linear attention architecture with a 1M-token context window. The result: lightning-fast responses, powerful reasoning, and outstanding cost efficiency. 🎁 Still 100% FREE on From high-frequency API calls and coding to complex Agents and long-document processing, jump in and experience the #1 model on for yourself! 👉 Try it free now:
0:13

🔥 GLM-5.3-Flash Hits #1 on 🐮 GLM-5.3-Flash (Ox Alpha) is now the most-used and most popular model on with cumulative token throughput surpassing 2.41 trillion tokens. As the first native omni-modal model in the GLM-5 series, it packs 320B total parameters, with 18B active, and features a hybrid sparse and linear attention architecture with a 1M-token context window. The result: lightning-fast responses, powerful reasoning, and outstanding cost efficiency. 🎁 Still 100% FREE on From high-frequency API calls and coding to complex Agents and long-document processing, jump in and experience the #1 model on for yourself! 👉 Try it free now:

B.AI

222,739 views • 1 month ago

most of you don't know how hard hermes agent is optimized for local AI at the system level. watch the full setup flow on screen. you paste an openai-compatible v1 endpoint, hermes auto-detects every model running behind it. doesn't matter if it's llama.cpp or vllm or any compatible server, all your models surface and become selectable in seconds. no config gymnastics, no manual model list. then it goes deeper. hermes ships with per model parsers, prompt template auto-handling, tool call format detection per model architecture, thinking mode awareness, all the small friction points other harnesses leak on. these were not built for cloud apis with one canonical model. they were built for builders running 10 different local models across 10 different stacks. cloud first harnesses bolt local support on top. hermes agent is local first from the architecture out. that's the system level gap. if you're getting started on local AI, this is the harness you start with. try for yourself and find out. anyone serious about local AI lands here eventually.
0:15

most of you don't know how hard hermes agent is optimized for local AI at the system level. watch the full setup flow on screen. you paste an openai-compatible v1 endpoint, hermes auto-detects every model running behind it. doesn't matter if it's llama.cpp or vllm or any compatible server, all your models surface and become selectable in seconds. no config gymnastics, no manual model list. then it goes deeper. hermes ships with per model parsers, prompt template auto-handling, tool call format detection per model architecture, thinking mode awareness, all the small friction points other harnesses leak on. these were not built for cloud apis with one canonical model. they were built for builders running 10 different local models across 10 different stacks. cloud first harnesses bolt local support on top. hermes agent is local first from the architecture out. that's the system level gap. if you're getting started on local AI, this is the harness you start with. try for yourself and find out. anyone serious about local AI lands here eventually.

Sudo su

14,407 views • 5 months ago

Google just dropped EmbeddingGemma 2. Open multimodal embedding model. 740M params. Runs fully on-device. Private. No cloud needed. 3 wild things you can do: 1. Find exact video moments with your voice
0:16

Google just dropped EmbeddingGemma 2. Open multimodal embedding model. 740M params. Runs fully on-device. Private. No cloud needed. 3 wild things you can do: 1. Find exact video moments with your voice

Min Choi

89,472 views • 7 days ago

Go local: ollama is here in Raycast 🦙 The power of Raycast AI, fully offline! Perfect for coding on flights or going completely incognito. Local models are free for everyone. No subscription needed.
0:18

Go local: ollama is here in Raycast 🦙 The power of Raycast AI, fully offline! Perfect for coding on flights or going completely incognito. Local models are free for everyone. No subscription needed.

Thomas Paul Mann

54,015 views • 1 year ago

cobalt 11 brings on-device media processing, huge performance improvements, and refreshed ui local processing is enabled for all desktop users starting now, so you can enjoy best compatibility with players and editing tools, no extra remuxing required :)
0:13

cobalt 11 brings on-device media processing, huge performance improvements, and refreshed ui local processing is enabled for all desktop users starting now, so you can enjoy best compatibility with players and editing tools, no extra remuxing required :)

cobalt

579,840 views • 1 year ago

Sonnet as the main model, Opus as an advisor, and Haiku for subagents Save this and try it with one of your sessions Install: npx claude-code-templates@latest --setting model/sonnet-opus-advisor-haiku-subagents This command installs the settings required to configure your Claude Code sessions with this model distribution Select the appropriate configuration scope: User, Project, or Local
0:17

Sonnet as the main model, Opus as an advisor, and Haiku for subagents Save this and try it with one of your sessions Install: npx claude-code-templates@latest --setting model/sonnet-opus-advisor-haiku-subagents This command installs the settings required to configure your Claude Code sessions with this model distribution Select the appropriate configuration scope: User, Project, or Local

Daniel San

28,256 views • 4 days ago

You can now route each request to the "fastest" ⚡ or "cheapest" 💸 provider with Hugging Face Inference Providers in Copilot Chat on Visual Studio Code!🤗 For each model, you can explicitly choose "cheapest" when price matters - or "fastest" when throughput is the priority. thanks to Simon Brandeis for shipping this!
0:18

You can now route each request to the "fastest" ⚡ or "cheapest" 💸 provider with Hugging Face Inference Providers in Copilot Chat on Visual Studio Code!🤗 For each model, you can explicitly choose "cheapest" when price matters - or "fastest" when throughput is the priority. thanks to Simon Brandeis for shipping this!

célina

33,234 views • 11 months ago

🔥 Introducing `npx agentlytics` a local, offline analytics dashboard for your AI editors. If you're switching between: you finally have one place to see all your sessions, tokens, tool calls, and model usage 👀 No cloud or sign-up required. →
0:13

🔥 Introducing `npx agentlytics` a local, offline analytics dashboard for your AI editors. If you're switching between: you finally have one place to see all your sessions, tokens, tool calls, and model usage 👀 No cloud or sign-up required. →

fatih kadir akın

37,863 views • 7 months ago

Today we’re releasing Code Llama, a large language model built on top of Llama 2, fine-tuned for coding & state-of-the-art for publicly available coding tools. Keeping with our open approach, Code Llama is publicly-available now for both research & commercial use. More ⬇️
0:14

Today we’re releasing Code Llama, a large language model built on top of Llama 2, fine-tuned for coding & state-of-the-art for publicly available coding tools. Keeping with our open approach, Code Llama is publicly-available now for both research & commercial use. More ⬇️

AI at Meta

1,005,204 views • 3 years ago

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

159,079 views • 6 months ago