Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

The MiniMax M2 model is mind-blowing! It's open-source. It outperforms Gemini 2.5, Claude 4.1, and Qwen3 across coding and tool-use benchmarks. Right now, it's one of the world's top 5 models in intelligence! And here is the best part: Claude is one of the best models you can use...

svpino's profile picture

Santiago

456,665 subscribers

91,263 Aufrufe • vor 1 Jahr •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

22 Kommentare

Profilbild von zeke
zekevor 1 Jahr

No downloadable weights no code just api access, how's this open source?

Profilbild von Chiyun Wang 王迟云
Chiyun Wang 王迟云vor 1 Jahr

Not convinced, I tried it extensively on lmarena and it's bad, always give the same template design, always mess the markdown. So either it's the lmarena version being trash

Profilbild von Saeed Anwar
Saeed Anwarvor 1 Jahr

MiniMax M2 beating Gemini 2.5 and Claude 4.1 while being open-source is the moment the game changed quietly. nobody talks about it because it's not American, but when your competitor's free model outperforms your paid flagship that's not a technical issue—that's a strategy issue

Profilbild von Bjorn Solstad - Devenia
Bjorn Solstad - Deveniavor 1 Jahr

I tried it, and it fucked up my Claude Code connection to claude ai. Not sure how to set it up in a clean way so they can coexist on my Linux Mint box. Any ideas?

Profilbild von Charles Waters
Charles Watersvor 11 Monaten

I see it on HuggingFace. Seems to require Ubuntu to run on. I'll download and experiment.

Profilbild von Hudpire
Hudpirevor 1 Jahr

is not working; it should be

Profilbild von Karan Jagtiani
Karan Jagtianivor 1 Jahr

Sounds impressive, especially with those benchmarks. Definitely on my radar now.

Profilbild von Rafiul
Rafiulvor 11 Monaten

GLM 4.6 still better

Profilbild von Jai Rod
Jai Rodvor 11 Monaten

Can you use minimax in Claude Code? Instead of sonnet? Or through a subagent?

Profilbild von Suhrab Khan⚡️
Suhrab Khan⚡️vor 11 Monaten

MiniMax M2 is insane! Ultra-efficient and perfect for multi-tool agent workflows. High performance at a fraction of the cost is a serious game-changer.

Profilbild von Andres Rey
Andres Reyvor 11 Monaten

The agentic capabilities are impressive for tool orchestration. For teams building autonomous workflows, this offers a solid cost-performance ratio worth testing.

Profilbild von POGINO-AI Life Coach for Your Goals
POGINO-AI Life Coach for Your Goalsvor 1 Jahr

That’s a huge leap for open-source AI. MiniMax M2 proves that intelligence, speed, and affordability can coexist. If it truly delivers agentic reliability at this scale, we’re watching the next evolution of accessible AGI.

Profilbild von Vansh Choudhary
Vansh Choudharyvor 11 Monaten

Shit; gotta go and try this right now. Something smells fishy though

Profilbild von Saïd Aitmbarek
Saïd Aitmbarekvor 11 Monaten

based share mate!

Profilbild von mzeid4real
mzeid4realvor 11 Monaten

It’s really amazing.

Profilbild von Diwakar Ray Yadav
Diwakar Ray Yadavvor 1 Jahr

Open source models catching up to the big players is exactly what we need to keep innovation moving forward. Top 5 is impressive

Profilbild von Lyceum
Lyceumvor 11 Monaten

Impressive insights! The MiniMax M2 truly sets a new standard in efficiency and performance. Excited to see its impact!

Profilbild von Kamorou Bate
Kamorou Batevor 1 Jahr

Saw it in my recommendation on Google but didn't have time to read it, but with all these I definitely have to check it out

Profilbild von gbdoin
gbdoinvor 11 Monaten

This post was brought to you by… MiniMax, the model that will allow Santiago to maximize his X account revenue with minimum effort. MiniMax!

Profilbild von Franco Robles 🇵🇭🇪🇸
Franco Robles 🇵🇭🇪🇸vor 11 Monaten

Has anyone tried this yet?

Profilbild von Dr. Tristan Behrens
Dr. Tristan Behrensvor 11 Monaten

@grok please find out: How many trainable parameters?

Profilbild von Kenshi
Kenshivor 1 Jahr

It’s rare to find so much capability in a lightweight model that still runs smooth.

Ähnliche Videos

MiniMax M2.1 is really good for writing code. It's open source, and you can connect Visual Studio Code to it and start coding right away. The model is especially good for those who don't use Python as a primary language. It's currently the best multilingual model out there. The model hits 72.5% on SWE-Multilingual and 88.6% on VIBE-Bench (best!) This means the model is also really good at performing real engineering work and does very well at agentic tasks: • Long tool chains • Structured agent scaffolding • Following repo-level rules and slash commands • Working inside existing agent frameworks without falling apart And obviously, it's faster, cheaper, and more concise than M2, which was already really good. I recorded a video to show you how to run MiniMax (official)'s 2.1 using Visual Studio Code.
5:45

MiniMax M2.1 is really good for writing code. It's open source, and you can connect Visual Studio Code to it and start coding right away. The model is especially good for those who don't use Python as a primary language. It's currently the best multilingual model out there. The model hits 72.5% on SWE-Multilingual and 88.6% on VIBE-Bench (best!) This means the model is also really good at performing real engineering work and does very well at agentic tasks: • Long tool chains • Structured agent scaffolding • Following repo-level rules and slash commands • Working inside existing agent frameworks without falling apart And obviously, it's faster, cheaper, and more concise than M2, which was already really good. I recorded a video to show you how to run MiniMax (official)'s 2.1 using Visual Studio Code.

Santiago

47,879 Aufrufe • vor 9 Monaten

You can use a 100% open source and MUCH cheaper/free alternative to Claude Code and Opus 4.5 OpenCode + MiniMax M2.1 can even build a 3D website using pure vibe coding. Steps are really simple: 1. Install OpenCode using the command 'npm i -g opencode-ai' 2. Get your MiniMax API key here: 3. Configure the MiniMax (official) mode Just type "opencode connect minimax" Coding plans start at $2… 10x cheaper than Claude Code (yes). You can also invite friends so they can get 10% off and you’ll get 10% API credits. (You can also use it locally if you have the config) And you're ready to build and iterate almost endlessly since the model is both way faster and cheaper than Opus in CC.
12:07

You can use a 100% open source and MUCH cheaper/free alternative to Claude Code and Opus 4.5 OpenCode + MiniMax M2.1 can even build a 3D website using pure vibe coding. Steps are really simple: 1. Install OpenCode using the command 'npm i -g opencode-ai' 2. Get your MiniMax API key here: 3. Configure the MiniMax (official) mode Just type "opencode connect minimax" Coding plans start at $2… 10x cheaper than Claude Code (yes). You can also invite friends so they can get 10% off and you’ll get 10% API credits. (You can also use it locally if you have the config) And you're ready to build and iterate almost endlessly since the model is both way faster and cheaper than Opus in CC.

Paul Couvert

42,443 Aufrufe • vor 9 Monaten

Want to use the latest cool AI models inside TRAE? You can now connect and run MiniMax-M2 (or any other OpenRouter model) in just minutes. This tutorial walks you through: - Setting up OpenRouter service provider - Adding your api key in TRAE - Selecting MiniMax-M2 in chat or coding mode — and building a 3D object with Three.js! Watch the full walkthrough ↓
3:21

Want to use the latest cool AI models inside TRAE? You can now connect and run MiniMax-M2 (or any other OpenRouter model) in just minutes. This tutorial walks you through: - Setting up OpenRouter service provider - Adding your api key in TRAE - Selecting MiniMax-M2 in chat or coding mode — and building a 3D object with Three.js! Watch the full walkthrough ↓

TRAE

16,445 Aufrufe • vor 11 Monaten

NOBODY wants to send their data to Google or OpenAI. Yet here we are, shipping proprietary code, customer information, and sensitive business logic to closed-source APIs we don't control. While everyone's chasing the latest closed-source releases, open-source models are quietly becoming the practical choice for many production systems. Here's what everyone is missing: Open-source models are catching up fast, and they bring something the big labs can't: privacy, speed, and control. I built a playground to test this myself. Used CometML's Opik to evaluate models on real code generation tasks - testing correctness, readability, and best practices against actual GitHub repos. Here's what surprised me: OSS models like MiniMax-M2, Kimi k2 performed on par with the likes of Gemini 3 and Claude Sonnet 4.5 on most tasks. But practically MiniMax-M2 turns out to be a winner as it's twice as fast and 12x cheaper when you compare it to models like Sonnet 4.5. Well, this isn't just about saving money. When your model is smaller and faster, you can deploy it in places closed-source APIs can't reach: ↳ Real-time applications that need sub-second responses ↳ Edge devices where latency kills user experience ↳ On-premise systems where data never leaves your infrastructure MiniMax-M2 runs with only 10B activated parameters. That efficiency means lower latency, higher throughput, and the ability to handle interactive agents without breaking the bank. The intelligence-to-cost ratio here changes what's possible. You're not choosing between quality and affordability anymore. You're not sacrificing privacy for performance. The gap is closing, and in many cases, it's already closed. If you're building anything that needs to be fast, private, or deployed at scale, it's worth taking a look at what's now available. MiniMax-M2 is 100% open-source, free for developers right now. I have shared the link to their GitHub repo in the next tweet. You will also find the code for the playground and evaluations I've done.
0:48

NOBODY wants to send their data to Google or OpenAI. Yet here we are, shipping proprietary code, customer information, and sensitive business logic to closed-source APIs we don't control. While everyone's chasing the latest closed-source releases, open-source models are quietly becoming the practical choice for many production systems. Here's what everyone is missing: Open-source models are catching up fast, and they bring something the big labs can't: privacy, speed, and control. I built a playground to test this myself. Used CometML's Opik to evaluate models on real code generation tasks - testing correctness, readability, and best practices against actual GitHub repos. Here's what surprised me: OSS models like MiniMax-M2, Kimi k2 performed on par with the likes of Gemini 3 and Claude Sonnet 4.5 on most tasks. But practically MiniMax-M2 turns out to be a winner as it's twice as fast and 12x cheaper when you compare it to models like Sonnet 4.5. Well, this isn't just about saving money. When your model is smaller and faster, you can deploy it in places closed-source APIs can't reach: ↳ Real-time applications that need sub-second responses ↳ Edge devices where latency kills user experience ↳ On-premise systems where data never leaves your infrastructure MiniMax-M2 runs with only 10B activated parameters. That efficiency means lower latency, higher throughput, and the ability to handle interactive agents without breaking the bank. The intelligence-to-cost ratio here changes what's possible. You're not choosing between quality and affordability anymore. You're not sacrificing privacy for performance. The gap is closing, and in many cases, it's already closed. If you're building anything that needs to be fast, private, or deployed at scale, it's worth taking a look at what's now available. MiniMax-M2 is 100% open-source, free for developers right now. I have shared the link to their GitHub repo in the next tweet. You will also find the code for the playground and evaluations I've done.

Akshay 🚀

50,323 Aufrufe • vor 11 Monaten

MiniMax M2.1 is officially live🚀 Built for real-world coding and AI-native organizations — from vibe builds to serious workflows. A SOTA 10B-activated OSS coding & agent model, scoring 72.5% on SWE-multilingual and 88.6% on our newly open-sourced VIBE-bench, exceeding leading closed-source models like Gemini 3 Pro and Claude 4.5 Sonnet. The most powerful OSS model for the agentic era is here.
0:54

MiniMax M2.1 is officially live🚀 Built for real-world coding and AI-native organizations — from vibe builds to serious workflows. A SOTA 10B-activated OSS coding & agent model, scoring 72.5% on SWE-multilingual and 88.6% on our newly open-sourced VIBE-bench, exceeding leading closed-source models like Gemini 3 Pro and Claude 4.5 Sonnet. The most powerful OSS model for the agentic era is here.

MiniMax (official)

1,074,732 Aufrufe • vor 10 Monaten

DeepSeek V4.1 Flash 1017 vs Haiku 5.5 vs MiniMax M3.1 Flash vs Muse Spark 1.3 You guys loved the Alpine benchmark with the frontier models, so here it is with the budget models. 🛩️ Honestly, DeepSeek V4.1 Flash 1017 blew me away. 🤯 It's 14x cheaper than Haiku and almost 3x faster, and it gave the best result of the four. MiniMax is right behind, and even cheaper at 23 cents. And Muse Spark did it for free! What do you guys think? 👀
0:25

DeepSeek V4.1 Flash 1017 vs Haiku 5.5 vs MiniMax M3.1 Flash vs Muse Spark 1.3 You guys loved the Alpine benchmark with the frontier models, so here it is with the budget models. 🛩️ Honestly, DeepSeek V4.1 Flash 1017 blew me away. 🤯 It's 14x cheaper than Haiku and almost 3x faster, and it gave the best result of the four. MiniMax is right behind, and even cheaper at 23 cents. And Muse Spark did it for free! What do you guys think? 👀

Tony

11,385 Aufrufe • vor 5 Tagen

I gave MiniMax M2.7 a task. It didn't just do it. It pushed back with questions. Been testing MiniMax M2.7 via Droid + MiniMax (official). Tool call for planning? Really good. The model keeps asking questions, and that makes the plan better. What I didn't expect, self-correction. When the plan and the execution drift, MiniMax notices. And fixes it. That's the kind of model behavior I want to see more of.
3:59

I gave MiniMax M2.7 a task. It didn't just do it. It pushed back with questions. Been testing MiniMax M2.7 via Droid + MiniMax (official). Tool call for planning? Really good. The model keeps asking questions, and that makes the plan better. What I didn't expect, self-correction. When the plan and the execution drift, MiniMax notices. And fixes it. That's the kind of model behavior I want to see more of.

λL-D1 | AI for Buzzer 🍉

28,044 Aufrufe • vor 7 Monaten

GPT-5.4 is here. Native computer-use capabilities. Up to 1M tokens of context in Codex and the API. Best-in-class agentic coding for complex tasks. Scalable tool search across larger ecosystems. More efficient reasoning for long, tool-heavy workflows.
3:02

GPT-5.4 is here. Native computer-use capabilities. Up to 1M tokens of context in Codex and the API. Best-in-class agentic coding for complex tasks. Scalable tool search across larger ecosystems. More efficient reasoning for long, tool-heavy workflows.

OpenAI Developers

1,076,232 Aufrufe • vor 7 Monaten

I've been testing Gaussian Splatting with MiniMax H3 locally, and all I can say is that MiniMax is currently the best model I've tried for creating GS models. The results are significantly better than what I get with LTX 2.
1:03

I've been testing Gaussian Splatting with MiniMax H3 locally, and all I can say is that MiniMax is currently the best model I've tried for creating GS models. The results are significantly better than what I get with LTX 2.

Alex

28,641 Aufrufe • vor 2 Monaten

Two AI models just became a cheat code. Claude 4.6 thinks. Minimax M2.5 executes. That’s the split. Claude: • 1M token context window • Reads entire codebases • Builds full strategies • Reviews complex work Minimax: • 20x cheaper than flagship models • 80%+ on real coding benchmarks • 37% faster than older versions • Built for agent workflows Here’s the play: 1. Claude plans the system. 2. Minimax builds it. 3. Claude reviews it. 4. Minimax refines it. Strategy → Execution → Review → Scale. That loop is unfair.
8:07

Two AI models just became a cheat code. Claude 4.6 thinks. Minimax M2.5 executes. That’s the split. Claude: • 1M token context window • Reads entire codebases • Builds full strategies • Reviews complex work Minimax: • 20x cheaper than flagship models • 80%+ on real coding benchmarks • 37% faster than older versions • Built for agent workflows Here’s the play: 1. Claude plans the system. 2. Minimax builds it. 3. Claude reviews it. 4. Minimax refines it. Strategy → Execution → Review → Scale. That loop is unfair.

Julian Goldie SEO

17,034 Aufrufe • vor 8 Monaten

Big day for open source. MiniMax Music 3 is undeniably a state of the art open weights Music Generation model and a real alternative to Suno. My favorite part about open weights: the best is yet to come. Once the community starts playing with the model and training LoRAs, this model only gets better and allows for more control.
1:59

Big day for open source. MiniMax Music 3 is undeniably a state of the art open weights Music Generation model and a real alternative to Suno. My favorite part about open weights: the best is yet to come. Once the community starts playing with the model and training LoRAs, this model only gets better and allows for more control.

rob - comfyui

323,651 Aufrufe • vor 2 Monaten

When will MiniMax have a Fable level model? Olive Song, research lead at MiniMax (official), joins The Bench to answer that question, plus M3, open source models closing the gap on Vibe Code Bench, and the truth about 996 culture. Full episode out now! 7:19 – Inside MiniMax's research culture: AI agents tracking papers since early last year 18:01- The headline of M3: native multimodality and one million tokens of context 21:10- The bottleneck of building models is people, not compute 22:48 – Does MiniMax work 996? 25:47 - Olive's prediction and the 30 point gap on Vibe Code Bench 32:02- Model capabilities Olive is most excited about 32:35- The lab Olive most respects outside of MiniMax 36:24- Will more labs go open source a year from now?
38:45

When will MiniMax have a Fable level model? Olive Song, research lead at MiniMax (official), joins The Bench to answer that question, plus M3, open source models closing the gap on Vibe Code Bench, and the truth about 996 culture. Full episode out now! 7:19 – Inside MiniMax's research culture: AI agents tracking papers since early last year 18:01- The headline of M3: native multimodality and one million tokens of context 21:10- The bottleneck of building models is people, not compute 22:48 – Does MiniMax work 996? 25:47 - Olive's prediction and the 30 point gap on Vibe Code Bench 32:02- Model capabilities Olive is most excited about 32:35- The lab Olive most respects outside of MiniMax 36:24- Will more labs go open source a year from now?

Vals AI

49,462 Aufrufe • vor 3 Monaten

MINIMAX OPEN SOURCED A CODING AGENT THAT LIVES IN YOUR TERMINAL its called MiniMax Code CLI. i gave it one prompt, 19 minutes later i had a working interactive 3d globe. → 249 countries → 14 animated flight arcs across 16 hubs → click a country and the camera flies to it → fresnel atmosphere, bloom, the full pass it did it in 9 phases at about 96 tok/s, committing after every one. Mcode is MIT and it is not locked to MiniMax models, you can bring your own key and point it at claude, gpt, or a local model. setup in your terminal: 1.⁠ ⁠curl -fsSL | bash 2.⁠ ⁠cd into the project and run mcode 3.⁠ ⁠run mcode init . first, that is the step that writes the AGENTS.md the AGENTS.md is doing more work here than the model. build commands, what not to touch, and a rule that it verifies before it calls anything done i will drop the repo below
1:26

MINIMAX OPEN SOURCED A CODING AGENT THAT LIVES IN YOUR TERMINAL its called MiniMax Code CLI. i gave it one prompt, 19 minutes later i had a working interactive 3d globe. → 249 countries → 14 animated flight arcs across 16 hubs → click a country and the camera flies to it → fresnel atmosphere, bloom, the full pass it did it in 9 phases at about 96 tok/s, committing after every one. Mcode is MIT and it is not locked to MiniMax models, you can bring your own key and point it at claude, gpt, or a local model. setup in your terminal: 1.⁠ ⁠curl -fsSL | bash 2.⁠ ⁠cd into the project and run mcode 3.⁠ ⁠run mcode init . first, that is the step that writes the AGENTS.md the AGENTS.md is doing more work here than the model. build commands, what not to touch, and a rule that it verifies before it calls anything done i will drop the repo below

Israfil

25,336 Aufrufe • vor 1 Monat

So Alibaba has just released a coding plan with all the best open source models You’ve access to Qwen 3.5-Plus, Kimi K2.5, MiniMax M2.5, GLM-5, etc. And you can plug it to: - Claude Code (as I did) - OpenClaw - Codex - Cline… While being waaay cheaper than the paid plan of Codex, Claude & Cie. More below
0:36

So Alibaba has just released a coding plan with all the best open source models You’ve access to Qwen 3.5-Plus, Kimi K2.5, MiniMax M2.5, GLM-5, etc. And you can plug it to: - Claude Code (as I did) - OpenClaw - Codex - Cline… While being waaay cheaper than the paid plan of Codex, Claude & Cie. More below

Paul Couvert

88,449 Aufrufe • vor 7 Monaten

Today, we're announcing that Comfy is the first and only commercial license distributor for MiniMax (official)'s generative media models. MiniMax open weight models have been downloaded through Comfy close to 20M times. Now you can license them for commercial work. → Full commercial rights to your outputs → Fine-tuning and LoRA training, fully allowed → Client and agency projects covered → MiniMax Audio & Music, not just video Professional and Enterprise plans, with Enterprise covering every H3 model version. Request a license with the link below today ⬇️
0:31

Today, we're announcing that Comfy is the first and only commercial license distributor for MiniMax (official)'s generative media models. MiniMax open weight models have been downloaded through Comfy close to 20M times. Now you can license them for commercial work. → Full commercial rights to your outputs → Fine-tuning and LoRA training, fully allowed → Client and agency projects covered → MiniMax Audio & Music, not just video Professional and Enterprise plans, with Enterprise covering every H3 model version. Request a license with the link below today ⬇️

ComfyUI

145,785 Aufrufe • vor 1 Monat

Mfw Minimax H3, the new Open Source Audiovisual model that is vastly superior to all the Frontier models, is also the first model in history to be trained on and remember The Will Stancil Show and now chooses to generate propaganda posters of Will Stancil all on its own.
1:17

Mfw Minimax H3, the new Open Source Audiovisual model that is vastly superior to all the Frontier models, is also the first model in history to be trained on and remember The Will Stancil Show and now chooses to generate propaganda posters of Will Stancil all on its own.

Emily Youcis

14,831 Aufrufe • vor 2 Monaten

Introducing MiniMax FastH3 on Reactor - the latest open-source model by MiniMax (official). Stream 720p with audio at $0.007/second with first frame image support, faster and cheaper than ever. Get started now by cloning our demo app:
1:23

Introducing MiniMax FastH3 on Reactor - the latest open-source model by MiniMax (official). Stream 720p with audio at $0.007/second with first frame image support, faster and cheaper than ever. Get started now by cloning our demo app:

reactor

778,813 Aufrufe • vor 1 Monat

BREAKING: Google Gemma just pointed their audience at the leaderboard that made Gemma 4 26B A4B 2x faster on Mac. Some of those gains came from people pointing Gemini 3.7 at it! Google's frontier model, in the hands of strangers, speeding up Google's open-weight model. The rest came from GPT-5.6, Claude, Qwen3 and MiniMax on the open board at Yukon. The team that shipped the model showing up to watch is new. That only happens with open weights. Open frontier research accelerates open intelligence.
0:20

BREAKING: Google Gemma just pointed their audience at the leaderboard that made Gemma 4 26B A4B 2x faster on Mac. Some of those gains came from people pointing Gemini 3.7 at it! Google's frontier model, in the hands of strangers, speeding up Google's open-weight model. The rest came from GPT-5.6, Claude, Qwen3 and MiniMax on the open board at Yukon. The team that shipped the model showing up to watch is new. That only happens with open weights. Open frontier research accelerates open intelligence.

Pratik Gandhi

15,891 Aufrufe • vor 1 Monat