Loading video...

Video Failed to Load

Go Home

Subagents running locally and simultaneously on MacBook Pro M5 with Codex CLI + LM Studio to review code and find bugs using Qwen 3.6 Powered by the updated MLX engine with batching in beta in the app The batching speed boost is noticeable

adrgrondin's profile picture

Adrien Grondin

8,157 subscribers

75,549 views • 3 months ago •via X (Twitter)

EducationHealth & WellnessScience & Technology
LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Video on LM Studio MLX v.1.8.5 + Codex + qwen3.6-35b-a3b 6bit on an M5 Max 128GB. Video is in normal speed to give you an idea of the experience. Command used: codex --oss -m qwen/qwen3.6-35b-a3b Local AI is becoming powerful 💪
1:18

Video on LM Studio MLX v.1.8.5 + Codex + qwen3.6-35b-a3b 6bit on an M5 Max 128GB. Video is in normal speed to give you an idea of the experience. Command used: codex --oss -m qwen/qwen3.6-35b-a3b Local AI is becoming powerful 💪

Ivan Fioravanti ᯅ

40,565 views • 3 months ago

I’ve been using the new MacBook Pro M5 since launch and it’s a beast, especially compared to my old M1! It’s good for running the smaller AI models locally but also everything you do is just so fast. Here’s Gemma 3 QAT 12B running with MLX:
0:35

I’ve been using the new MacBook Pro M5 since launch and it’s a beast, especially compared to my old M1! It’s good for running the smaller AI models locally but also everything you do is just so fast. Here’s Gemma 3 QAT 12B running with MLX:

Adrien Grondin

101,950 views • 9 months ago

🚨BREAKING: OpenAI's Codex is now completely FREE to run locally with Ollama. No API costs. No rate limits. 100% private on your machine. You can now use both the Codex App and Codex CLI with powerful open-source models like DeepSeek V4, Gemma 4, and Qwen 3.6. Here's how to set it up in minutes:
0:31

🚨BREAKING: OpenAI's Codex is now completely FREE to run locally with Ollama. No API costs. No rate limits. 100% private on your machine. You can now use both the Codex App and Codex CLI with powerful open-source models like DeepSeek V4, Gemma 4, and Qwen 3.6. Here's how to set it up in minutes:

WorldofAI

817,638 views • 3 months ago

The new Qwen 3.5 4B runs incredibly well on M5. The model is close to GPT-4o in benchmarks. Running fully on-device with MLX.
0:20

The new Qwen 3.5 4B runs incredibly well on M5. The model is close to GPT-4o in benchmarks. Running fully on-device with MLX.

Adrien Grondin

230,466 views • 6 months ago

A long time coming but new mlx-lm is here with better batching support in the server and Gemma 4. pip install -U mlx-lm Here is a video where a single M3 Ultra serves 5 opencode sessions with Gemma 4 26B that process ~130k tokens in ~1.5 minutes.
2:03

A long time coming but new mlx-lm is here with better batching support in the server and Gemma 4. pip install -U mlx-lm Here is a video where a single M3 Ultra serves 5 opencode sessions with Gemma 4 26B that process ~130k tokens in ~1.5 minutes.

Angelos Katharopoulos

66,202 views • 5 months ago

Part 02 -- Universal CLI Agent Server Claude Code, Codex CLI, Qwen Code, Gemini CLI, or anything---Use ANY terminal AI agents right in the browser with 1 click, just like web surfing. Reply to help with beta.
3:32

Part 02 -- Universal CLI Agent Server Claude Code, Codex CLI, Qwen Code, Gemini CLI, or anything---Use ANY terminal AI agents right in the browser with 1 click, just like web surfing. Reply to help with beta.

cocktail peanut

12,217 views • 10 months ago

Next mlx-vlm release will ship with continuous batching support on the server 🚀 What's coming: → Continuous batching — new requests join the active batch immediately, no waiting. Mixed image + text batches supported → OpenAI-compatible API — field-for-field match with mlx-lm, reasoning/content split for thinking models, tag-aware streaming → Multi-turn tool calling — full tool use support across streaming and non-streaming, works with Gemma4 and other templates → Vision feature caching — cache image embeddings across turns. Gemma4: 228x speedup, Qwen3.5: 23x on cache hit All running locally on Apple Silicon. Check our this demo running 4 concurrent requests (mixed image + text) to gemma-4-26B-A4B-IT by Google Gemma in bf16 using Pi + MLX-VLM server on my M3 Ultra. One of the requests ingests a 8K resolution image!
0:52

Next mlx-vlm release will ship with continuous batching support on the server 🚀 What's coming: → Continuous batching — new requests join the active batch immediately, no waiting. Mixed image + text batches supported → OpenAI-compatible API — field-for-field match with mlx-lm, reasoning/content split for thinking models, tag-aware streaming → Multi-turn tool calling — full tool use support across streaming and non-streaming, works with Gemma4 and other templates → Vision feature caching — cache image embeddings across turns. Gemma4: 228x speedup, Qwen3.5: 23x on cache hit All running locally on Apple Silicon. Check our this demo running 4 concurrent requests (mixed image + text) to gemma-4-26B-A4B-IT by Google Gemma in bf16 using Pi + MLX-VLM server on my M3 Ultra. One of the requests ingests a 8K resolution image!

Prince Canuma

82,349 views • 4 months ago

I’ve tried driving Qwen 3.6 on my MacBook Pro with a few different agent harnesses: Claude Code: too slow. It moves like a turtle. Codex: Qwen stopped in the middle of tasks, probably due to tool-call mismatches between them. Qwen Code: so far the best experience. Very smooth. Qwen Code sometimes even tells a joke while you’re waiting. This video is played at original speed so you can get a real feel for how Qwen works.
0:47

I’ve tried driving Qwen 3.6 on my MacBook Pro with a few different agent harnesses: Claude Code: too slow. It moves like a turtle. Codex: Qwen stopped in the middle of tasks, probably due to tool-call mismatches between them. Qwen Code: so far the best experience. Very smooth. Qwen Code sometimes even tells a joke while you’re waiting. This video is played at original speed so you can get a real feel for how Qwen works.

Michael Guo

20,505 views • 3 months ago

Excited to release v1 of /review in the Codex CLI! You can now use gpt-5-codex to investigate and find critical bugs in your code. Check it out and let us know if you have any feedback!
0:25

Excited to release v1 of /review in the Codex CLI! You can now use gpt-5-codex to investigate and find critical bugs in your code. Check it out and let us know if you have any feedback!

daniel

743,130 views • 11 months ago

The latest Qwen 3 VL by Qwen running on iPhone 17 Pro with MLX Qwen 3 VL brings upgraded visual understanding, recognition, and OCR capabilities without sacrificing text performance like previous models The 4B model here is close to Qwen 2.5 VL 72B in many benchmarks
0:21

The latest Qwen 3 VL by Qwen running on iPhone 17 Pro with MLX Qwen 3 VL brings upgraded visual understanding, recognition, and OCR capabilities without sacrificing text performance like previous models The 4B model here is close to Qwen 2.5 VL 72B in many benchmarks

Adrien Grondin

109,700 views • 10 months ago

Here is a real-life run, end-to-end from training to using the trained LLM in LM Studio by LM Studio MLX-LoRA-Studio gives you a user friendly, extremely customisable training interface for fine-tuning and training LLMs. Its a MacOS native app, front- and backend. Will be releasing the App in the coming days. GH Repo:
4:41

Here is a real-life run, end-to-end from training to using the trained LLM in LM Studio by LM Studio MLX-LoRA-Studio gives you a user friendly, extremely customisable training interface for fine-tuning and training LLMs. Its a MacOS native app, front- and backend. Will be releasing the App in the coming days. GH Repo:

Gökdeniz Gülmez

11,952 views • 2 months ago

The new Qwen 3.5 by Qwen running on-device on iPhone 17 Pro. Qwen 3.5 beats models 4 times its size, has strong visual understanding, and can toggle reasoning on or off. The 2B 6-bit model here is running with MLX optimized for Apple Silicon.
0:25

The new Qwen 3.5 by Qwen running on-device on iPhone 17 Pro. Qwen 3.5 beats models 4 times its size, has strong visual understanding, and can toggle reasoning on or off. The 2B 6-bit model here is running with MLX optimized for Apple Silicon.

Adrien Grondin

3,558,786 views • 6 months ago

Gemma 4 is now in Android Studio! You can use Android Studio Agent mode to develop features, vibe code Android apps, refactor your code, and fix bugs, all locally, and powered by Gemma!
0:20

Gemma 4 is now in Android Studio! You can use Android Studio Agent mode to develop features, vibe code Android apps, refactor your code, and fix bugs, all locally, and powered by Gemma!

Omar Sanseviero

36,927 views • 5 months ago

Your agent works out the route. You review and sign every transaction. A new way to exit DeFi positions powered by Zerion CLI: → Find a position in Zerion → Click “Withdraw with Zerion AI” → Run the prompt in Claude Code → Review and sign in Zerion Watch the demo ↓
1:20

Your agent works out the route. You review and sign every transaction. A new way to exit DeFi positions powered by Zerion CLI: → Find a position in Zerion → Click “Withdraw with Zerion AI” → Run the prompt in Claude Code → Review and sign in Zerion Watch the demo ↓

Zerion

29,460 views • 1 month ago

OpenAI Codex is now integrated directly in Visual Studio Code through the new Agent Sessions view - and can be powered by your GitHub Copilot subscription. Try it out now with VS Code Insiders and a Copilot Pro+ subscription. Happy coding!
0:32

OpenAI Codex is now integrated directly in Visual Studio Code through the new Agent Sessions view - and can be powered by your GitHub Copilot subscription. Try it out now with VS Code Insiders and a Copilot Pro+ subscription. Happy coding!

Visual Studio Code

335,103 views • 10 months ago

Your Mac is about to run inference like a datacenter. Coming soon to MLX-Swift: Continuous batching: the fastest way to handle multiple inference streams locally. It starts with regular inference and seamlessly upgrades to batched mode when new requests arrive. The best of both worlds. Based on the work of and Awni Hannun
0:25

Your Mac is about to run inference like a datacenter. Coming soon to MLX-Swift: Continuous batching: the fastest way to handle multiple inference streams locally. It starts with regular inference and seamlessly upgrades to batched mode when new requests arrive. The best of both worlds. Based on the work of and Awni Hannun

Ronald Mannak

60,570 views • 10 months ago

The top question I got was does Halo support local LLM running on a Mac. Here's an example of the app using gemma-4-26b-a4b-qat via LM Studio running on a base M5/32GB. It's giving an average of 40 tok/s and in my brief testing is able to execute all tool calls correctly. Very usable with this app.
0:37

The top question I got was does Halo support local LLM running on a Mac. Here's an example of the app using gemma-4-26b-a4b-qat via LM Studio running on a base M5/32GB. It's giving an average of 40 tok/s and in my brief testing is able to execute all tool calls correctly. Very usable with this app.

Tanmay

11,682 views • 2 months ago

CLI coding agents now have first-class support in Warp: • Built-in voice transcription powered by Wispr Flow • Attach images to your prompt easily • Browse files and review code directly in Warp Works with Claude Code, Codex, Amp, Gemini, and Droid 🙌
0:45

CLI coding agents now have first-class support in Warp: • Built-in voice transcription powered by Wispr Flow • Attach images to your prompt easily • Browse files and review code directly in Warp Works with Claude Code, Codex, Amp, Gemini, and Droid 🙌

Warp

122,432 views • 7 months ago