Video yükleniyor...

Video Yüklenemedi

Ana Sayfaya Dön

Running four simultaneous OpenCode agents works well with mlx_lm.server continuous batching and MiniMax M2.1 on an M3 Ultra:

awnihannun's profile picture

Awni Hannun

44,897 subscribers

95,092 görüntüleme • 8 ay önce •via X (Twitter)

Haberler & PolitikaSağlık & İyilik
LIVE

Anya Rossi• Live Now

Private livecam show

0 Yorum

Yorum bulunmuyor

Orijinal gönderinin yorumları burada görünecek

Benzer Videolar

Running Minimax M2.1 (MiniMax (official)) with OpenCode (OpenCode) and mlx_lm.server. Works quite well on an M3 Ultra. Once the KV cache is warm the prompt processing is pretty quick. And token generation is very fast.
0:41

Running Minimax M2.1 (MiniMax (official)) with OpenCode (OpenCode) and mlx_lm.server. Works quite well on an M3 Ultra. Once the KV cache is warm the prompt processing is pretty quick. And token generation is very fast.

Awni Hannun

32,329 görüntüleme • 8 ay önce

Running four high-level OpenCode agents + subagents with mlx_lm.server continuous batching and MiniMax M2.5 (6-bit). Fits easily on a 512GB M3 Ultra. Generation is quite fast. But prefill is still slow compared to cloud servers.
0:26

Running four high-level OpenCode agents + subagents with mlx_lm.server continuous batching and MiniMax M2.5 (6-bit). Fits easily on a 512GB M3 Ultra. Generation is quite fast. But prefill is still slow compared to cloud servers.

Awni Hannun

25,535 görüntüleme • 6 ay önce

MiniMax M2.1 in 4-bit cruises on an M3 Ultra with mlx-lm. Generated a space invaders game using 5098 tokens at 47.2 tok/sec:
0:23

MiniMax M2.1 in 4-bit cruises on an M3 Ultra with mlx-lm. Generated a space invaders game using 5098 tokens at 47.2 tok/sec:

Awni Hannun

94,109 görüntüleme • 8 ay önce

I wasn't expecting such an intelligence model on my local machine. Minimax 2.1 on MLX🔥 git clone cd mlx-lm && pip install -e . mlx_lm.server --model mlx-community/MiniMax-M2.1-4bit
1:14

I wasn't expecting such an intelligence model on my local machine. Minimax 2.1 on MLX🔥 git clone cd mlx-lm && pip install -e . mlx_lm.server --model mlx-community/MiniMax-M2.1-4bit

mzba

30,377 görüntüleme • 8 ay önce

I did it! It works! Using GLM-4.7-4bit with mlx_lm.server and opencode to fix real code locally! 🔥 Here single M3 Ultra 512GB, nex step phase will be 2 using Tensor Parallelism and then apply same changes to exo. Prefill is slow on a single machine, but generation is good.
1:44

I did it! It works! Using GLM-4.7-4bit with mlx_lm.server and opencode to fix real code locally! 🔥 Here single M3 Ultra 512GB, nex step phase will be 2 using Tensor Parallelism and then apply same changes to exo. Prefill is slow on a single machine, but generation is good.

Ivan Fioravanti ᯅ

44,000 görüntüleme • 8 ay önce

MiniMax M3 support added to mlx-vlm with MSA implementation! 🚀 Tested on M3 Ultra 512GB running at 24 tps with peak memory ~240GB. Now working on optimizing performance and adding ton of tests 💪 Model is here: PR is here:
0:24

MiniMax M3 support added to mlx-vlm with MSA implementation! 🚀 Tested on M3 Ultra 512GB running at 24 tps with peak memory ~240GB. Now working on optimizing performance and adding ton of tests 💪 Model is here: PR is here:

Ivan Fioravanti ᯅ

24,376 görüntüleme • 2 ay önce

OpenCode + GLM-4.7-Flash 8bit. llama-server vs mlx_lm.server. Using one M3 Ultra for each host. Same prompts, you can check duration for each step in the video. Amazing experience on both side! Local coding AI model are becoming reality! 🔥
2:21

OpenCode + GLM-4.7-Flash 8bit. llama-server vs mlx_lm.server. Using one M3 Ultra for each host. Same prompts, you can check duration for each step in the video. Amazing experience on both side! Local coding AI model are becoming reality! 🔥

Ivan Fioravanti ᯅ

50,771 görüntüleme • 7 ay önce

This is MiniMax-M2.5 MLX running in LM Studio on an Apple Mac Studio M3 Ultra 512GB. Fast enough out of the box for hosting OpenClaw, n8n workflows, and Open WebUI for the team.
0:53

This is MiniMax-M2.5 MLX running in LM Studio on an Apple Mac Studio M3 Ultra 512GB. Fast enough out of the box for hosting OpenClaw, n8n workflows, and Open WebUI for the team.

Patrick J Kennedy

76,574 görüntüleme • 6 ay önce

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 görüntüleme • 5 ay önce

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 görüntüleme • 4 ay önce

A perfect coding model for MLX on Apple silicon.. Qwen delivered again. Runs quite fast on an M3 Ultra. Running the 4-bit quantized with mlx-lm:
0:50

A perfect coding model for MLX on Apple silicon.. Qwen delivered again. Runs quite fast on an M3 Ultra. Running the 4-bit quantized with mlx-lm:

Awni Hannun

186,772 görüntüleme • 1 yıl önce

MiniMax M2.1 vs Opus 4.5 vs GLM-4.7 built an interactive 3D solar system from scratch. Running on 3 Claude Code in parallel. The winner is: 🥇 GLM-4.7! 🥈 Opus 4.5 (slowest) 🥉 M2.1 (fastest) GLM-4.7 design capabilities are another level vs 4.6!
0:48

MiniMax M2.1 vs Opus 4.5 vs GLM-4.7 built an interactive 3D solar system from scratch. Running on 3 Claude Code in parallel. The winner is: 🥇 GLM-4.7! 🥈 Opus 4.5 (slowest) 🥉 M2.1 (fastest) GLM-4.7 design capabilities are another level vs 4.6!

Ivan Fioravanti ᯅ

81,630 görüntüleme • 8 ay önce

We ran the same prompt and identical starting context on MiniMax-M2.1 and Claude Sonnet 4.5. MiniMax-M2.1 by MiniMax (official) reached a usable result faster, required fewer structural fixes, and produced a more consistent visual and interaction flow from the first pass. To test this properly, we asked both models to build a complex single-page web animation with real-world visual and physics constraints. Comparison video below 👇
0:36

We ran the same prompt and identical starting context on MiniMax-M2.1 and Claude Sonnet 4.5. MiniMax-M2.1 by MiniMax (official) reached a usable result faster, required fewer structural fixes, and produced a more consistent visual and interaction flow from the first pass. To test this properly, we asked both models to build a complex single-page web animation with real-world visual and physics constraints. Comparison video below 👇

GitHub Projects Community

29,335 görüntüleme • 7 ay önce

ds4f-mxfp4 branch running locally on a single M3 Ultra at ~31 toks/s building pagoda and frogger tests. 🚀
0:23

ds4f-mxfp4 branch running locally on a single M3 Ultra at ~31 toks/s building pagoda and frogger tests. 🚀

Ivan Fioravanti ᯅ

21,089 görüntüleme • 1 ay önce

GLM-4.7-8bit (350GB) running at 19 toks/s on two M3 Ultra 512GB using Tensor Parallelism with EXO - MLX, versus 14 toks/s with single node. 🚀 Now context benchmarking & then OpenCode tests 🔥 Note: this is from sources, I had to change things to run it.
0:50

GLM-4.7-8bit (350GB) running at 19 toks/s on two M3 Ultra 512GB using Tensor Parallelism with EXO - MLX, versus 14 toks/s with single node. 🚀 Now context benchmarking & then OpenCode tests 🔥 Note: this is from sources, I had to change things to run it.

Ivan Fioravanti ᯅ

327,687 görüntüleme • 8 ay önce

Win 3 months of MiniMax Token Plan Max and $200 in GMI credits! We're running a 14-day MiniMaxthon with MiniMax (official) , one winner per track: - Multimodal - Synthesis (Multimodal + LLM) - Reasoning (LLM only) Build for free using M3, M2.7, Music 3.0, and Speech 2.8 on GMI Submit 👇
0:35

Win 3 months of MiniMax Token Plan Max and $200 in GMI credits! We're running a 14-day MiniMaxthon with MiniMax (official) , one winner per track: - Multimodal - Synthesis (Multimodal + LLM) - Reasoning (LLM only) Build for free using M3, M2.7, Music 3.0, and Speech 2.8 on GMI Submit 👇

GMI Cloud

35,220 görüntüleme • 12 gün önce

Qwen 3.5 397B prompt processing on M3 Ultra (with MLX distributed + JACCL) - 3.4× speedup on 4 chips - scaling improves as context increases Really fun to use with opencode; generated a playable Asteroids clone in ~4 minutes (real time, including me playing it a bit).
4:13

Qwen 3.5 397B prompt processing on M3 Ultra (with MLX distributed + JACCL) - 3.4× speedup on 4 chips - scaling improves as context increases Really fun to use with opencode; generated a playable Asteroids clone in ~4 minutes (real time, including me playing it a bit).

Angelos Katharopoulos

23,552 görüntüleme • 6 ay önce

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
1:40

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

Adrien Grondin

75,549 görüntüleme • 3 ay önce

So I created a simple Expo app for remotely connecting to an OpenCode server (running on my mac) so I can remotely control & prompt from my iPad or phone. Not very polished but it works pretty well and its kinda interesting that you can see the feedback loop live.
1:35

So I created a simple Expo app for remotely connecting to an OpenCode server (running on my mac) so I can remotely control & prompt from my iPad or phone. Not very polished but it works pretty well and its kinda interesting that you can see the feedback loop live.

ryan vogel

50,706 görüntüleme • 8 ay önce

What is this Nanbeige4.1-3B model running at - 77 toks/s in bf16 (in video) - 115 toks/s in 8bit on M3 Ultra with MLX with these benchmark scores! 🔥
0:27

What is this Nanbeige4.1-3B model running at - 77 toks/s in bf16 (in video) - 115 toks/s in 8bit on M3 Ultra with MLX with these benchmark scores! 🔥

Ivan Fioravanti ᯅ

85,228 görüntüleme • 6 ay önce