×
I can’t help rooting for tiny open source AI model maker Arcee | TechCrunch
Arcee, a tiny 26-person U.S. startup that built a massive, 400B-parameter open source LLM on a  million shoestring budget, has released its new reasoning model. Arcee calls the model Trinity Large Thinking — and it’s the most capable open-weight model “ever released by a non-Chinese company,” claims CEO Mark McQuade to TechCrunch.

As that comment implies, Arcee has a goal that I can’t help but root for: It wants to give U.S. and Western companies a model that gives them no reason to use a Chinese-based one.







While Chinese models are extremely capable, they are perceived as risky, putting power, and perhaps data, into the hands of a government that doesn’t share all of the Western world’s ideals.

With Arcee, companies can download the model, train it to their own needs, and use it on premises. Companies can also use Arcee’s cloud-hosted version, accessible via API.

While Arcee’s models are not outperforming the closed source models from the big labs like Anthropic or OpenAI, they’re not being held hostage by the whims of those giants, either. 

For instance, Claude, with its exceptional abilities to code, has been a popular choice for users of open source AI agent tool OpenClaw. But Anthropic pulled the rug out from them last week when it told users that their Anthropic subscriptions will no longer cover OpenClaw usage — they will have to pay additionally for that. (In February, OpenClaw creator Peter Steinberger said he was joining Anthropic’s biggest rival, OpenAI.)

In contrast, McQuade proudly points to data from OpenRouter that says it has become one of the top models used with OpenClaw.

	
		
		Techcrunch event
		
			
			
									San Francisco, CA
													|
													October 13-15, 2026
							
			
		
	


So, how good is Trinity Large Thinking? It is comparable to some of the other top open source models, according to the benchmark results it shared with TechCrunch.

Arcee Trinity large thinking BenchmarksImage Credits:Arcee / Arcee

As we previously reported, it is not a head-to-head threat to the big cheese among U.S.-built open models: Meta’s Llama 4. But it also doesn’t have the odd, not-really open source license issues of Meta’s model. All of Arcee’s Trinity models are released under the gold standard for OS licenses, Apache 2.0.

Just to be clear, there are also countless other U.S. startups offering open source models and, as a fan of the ingenuity of startups, I’m rooting for them, too.








#rooting #tiny #open #source #model #maker #Arcee #TechCrunchArcee AI,chinese ai,open source ai

I can’t help rooting for tiny open source AI model maker Arcee | TechCrunch

Arcee, a tiny 26-person U.S. startup that built a massive, 400B-parameter open source LLM on a $20 million shoestring budget, has released its new reasoning model. Arcee calls the model Trinity Large Thinking — and it’s the most capable open-weight model “ever released by a non-Chinese company,” claims CEO Mark McQuade to TechCrunch.

As that comment implies, Arcee has a goal that I can’t help but root for: It wants to give U.S. and Western companies a model that gives them no reason to use a Chinese-based one.

While Chinese models are extremely capable, they are perceived as risky, putting power, and perhaps data, into the hands of a government that doesn’t share all of the Western world’s ideals.

With Arcee, companies can download the model, train it to their own needs, and use it on premises. Companies can also use Arcee’s cloud-hosted version, accessible via API.

While Arcee’s models are not outperforming the closed source models from the big labs like Anthropic or OpenAI, they’re not being held hostage by the whims of those giants, either.

For instance, Claude, with its exceptional abilities to code, has been a popular choice for users of open source AI agent tool OpenClaw. But Anthropic pulled the rug out from them last week when it told users that their Anthropic subscriptions will no longer cover OpenClaw usage — they will have to pay additionally for that. (In February, OpenClaw creator Peter Steinberger said he was joining Anthropic’s biggest rival, OpenAI.)

In contrast, McQuade proudly points to data from OpenRouter that says it has become one of the top models used with OpenClaw.

Techcrunch event

San Francisco, CA | October 13-15, 2026

So, how good is Trinity Large Thinking? It is comparable to some of the other top open source models, according to the benchmark results it shared with TechCrunch.

I can’t help rooting for tiny open source AI model maker Arcee | TechCrunch
Arcee, a tiny 26-person U.S. startup that built a massive, 400B-parameter open source LLM on a  million shoestring budget, has released its new reasoning model. Arcee calls the model Trinity Large Thinking — and it’s the most capable open-weight model “ever released by a non-Chinese company,” claims CEO Mark McQuade to TechCrunch.

As that comment implies, Arcee has a goal that I can’t help but root for: It wants to give U.S. and Western companies a model that gives them no reason to use a Chinese-based one.







While Chinese models are extremely capable, they are perceived as risky, putting power, and perhaps data, into the hands of a government that doesn’t share all of the Western world’s ideals.

With Arcee, companies can download the model, train it to their own needs, and use it on premises. Companies can also use Arcee’s cloud-hosted version, accessible via API.

While Arcee’s models are not outperforming the closed source models from the big labs like Anthropic or OpenAI, they’re not being held hostage by the whims of those giants, either. 

For instance, Claude, with its exceptional abilities to code, has been a popular choice for users of open source AI agent tool OpenClaw. But Anthropic pulled the rug out from them last week when it told users that their Anthropic subscriptions will no longer cover OpenClaw usage — they will have to pay additionally for that. (In February, OpenClaw creator Peter Steinberger said he was joining Anthropic’s biggest rival, OpenAI.)

In contrast, McQuade proudly points to data from OpenRouter that says it has become one of the top models used with OpenClaw.

	
		
		Techcrunch event
		
			
			
									San Francisco, CA
													|
													October 13-15, 2026
							
			
		
	


So, how good is Trinity Large Thinking? It is comparable to some of the other top open source models, according to the benchmark results it shared with TechCrunch.

Arcee Trinity large thinking BenchmarksImage Credits:Arcee / Arcee

As we previously reported, it is not a head-to-head threat to the big cheese among U.S.-built open models: Meta’s Llama 4. But it also doesn’t have the odd, not-really open source license issues of Meta’s model. All of Arcee’s Trinity models are released under the gold standard for OS licenses, Apache 2.0.

Just to be clear, there are also countless other U.S. startups offering open source models and, as a fan of the ingenuity of startups, I’m rooting for them, too.








#rooting #tiny #open #source #model #maker #Arcee #TechCrunchArcee AI,chinese ai,open source ai
Arcee Trinity large thinking BenchmarksImage Credits:Arcee / Arcee

As we previously reported, it is not a head-to-head threat to the big cheese among U.S.-built open models: Meta’s Llama 4. But it also doesn’t have the odd, not-really open source license issues of Meta’s model. All of Arcee’s Trinity models are released under the gold standard for OS licenses, Apache 2.0.

Just to be clear, there are also countless other U.S. startups offering open source models and, as a fan of the ingenuity of startups, I’m rooting for them, too.

#rooting #tiny #open #source #model #maker #Arcee #TechCrunchArcee AI,chinese ai,open source ai

Arcee, a tiny 26-person U.S. startup that built a massive, 400B-parameter open source LLM on a $20 million shoestring budget, has released its new reasoning model. Arcee calls the model Trinity Large Thinking — and it’s the most capable open-weight model “ever released by a non-Chinese company,” claims CEO Mark McQuade to TechCrunch.

As that comment implies, Arcee has a goal that I can’t help but root for: It wants to give U.S. and Western companies a model that gives them no reason to use a Chinese-based one.

While Chinese models are extremely capable, they are perceived as risky, putting power, and perhaps data, into the hands of a government that doesn’t share all of the Western world’s ideals.

With Arcee, companies can download the model, train it to their own needs, and use it on premises. Companies can also use Arcee’s cloud-hosted version, accessible via API.

While Arcee’s models are not outperforming the closed source models from the big labs like Anthropic or OpenAI, they’re not being held hostage by the whims of those giants, either.

For instance, Claude, with its exceptional abilities to code, has been a popular choice for users of open source AI agent tool OpenClaw. But Anthropic pulled the rug out from them last week when it told users that their Anthropic subscriptions will no longer cover OpenClaw usage — they will have to pay additionally for that. (In February, OpenClaw creator Peter Steinberger said he was joining Anthropic’s biggest rival, OpenAI.)

In contrast, McQuade proudly points to data from OpenRouter that says it has become one of the top models used with OpenClaw.

Techcrunch event

San Francisco, CA
|
October 13-15, 2026

So, how good is Trinity Large Thinking? It is comparable to some of the other top open source models, according to the benchmark results it shared with TechCrunch.

Arcee Trinity large thinking BenchmarksImage Credits:Arcee / Arcee

As we previously reported, it is not a head-to-head threat to the big cheese among U.S.-built open models: Meta’s Llama 4. But it also doesn’t have the odd, not-really open source license issues of Meta’s model. All of Arcee’s Trinity models are released under the gold standard for OS licenses, Apache 2.0.

Just to be clear, there are also countless other U.S. startups offering open source models and, as a fan of the ingenuity of startups, I’m rooting for them, too.

Source link
#rooting #tiny #open #source #model #maker #Arcee #TechCrunch

Previous post

Daredevil: Born Again’s Huge ‘Death’ Clarified by Vincent D’Onofrio & EPs

Next post

Lakers vs Oklahoma City — Why is LeBron James not playing in NBA today? <div id="content-body-70836919" itemprop="articleBody"><p>LeBron James is sitting out the Los Angeles Lakers ‘ game against Oklahoma City Thunder on Tuesday night to rest his injured left foot.</p><p>Lakers are already without Luka Doncic and Austin Reaves, who are both sidelined for at least the rest of the regular season with significant injuries. Doncic, Reaves and James are averaging a combined 77.6 points per game this season, or 66.5 per cent of the Lakers’ typical nightly scoring output.</p><p>The Lakers also are without starting guard Marcus Smart (bruised right ankle) and back-up centre Jaxson Hayes (bruised left foot) against the NBA-leading Thunder.</p><p>The 41-year-old James has been dealing with left foot arthritis and discomfort for much of the regular season, and he has occasionally sat out games to rest. The visit from the Thunder will be the 22nd game he has missed in his unprecedented 23rd NBA season.</p><p>Although James’ absence is only a short-term decision to rest a veteran in a game that would be difficult to win with a full roster, the Lakers’ injury woes in the past week have badly damaged their hopes of being a championship contender in Doncic’s first full season with the club.</p><p>Both Doncic — the NBA’s scoring leader — and Reaves were injured during the Lakers’ blowout loss to the Thunder last week. Doncic incurred a Grade 2 strain of his left hamstring, and Reaves injured his oblique muscles.</p><p>The Lakers have indicated only that both Doncic and Reaves will miss the rest of this week, but their injuries typically take longer than that to heal in most athletes. Coach JJ Redick seemed to acknowledge Tuesday they’ll be significantly short-handed when the playoffs begin.</p><p>“Our playoff rotation is not set in the first round with AR and Luka being out, so we’ve got to evaluate everybody and we’ve got to find who those guys are that we can trust and put in the playoffs,” Redick said.</p><p>Doncic has left the Lakers to travel to Europe for treatment in an attempt to shorten his recovery time. Redick wouldn’t say how long Doncic is likely to be away from the team.</p><p>“I think he’s in good spirits,” Redick said. “In my conversations with him, he’s motivated to do everything possible. I know for him, it’s hard not for him to be on a basketball court. That’s his happy place, and he’s one of the handful of guys that really played year-round. … He wants to get back on the court.”</p><p><b>ALSO READ | <a href="https://sportstar.thehindu.com/basketball/victor-wembanyama-injury-update-san-antonio-spurs-star-doubtful-for-rest-of-nba-season/article70832984.ece" target="_blank" rel="nofollow">Victor Wembanyama injury update — San Antonio Spurs star doubtful for rest of NBA season</a></b></p><p>In Lakers’ first game after losing Doncic and Reaves last Sunday, James had 30 points, 15 assists and nine rebounds, but the Lakers lost 134-128 to the struggling Dallas Mavericks.</p><p>James is averaging 20.8 points, 7.1 assists and 6.1 rebounds per game this season. He has knocked over another set of NBA records this season, most recently surpassing Robert Parish’s record for career games played.</p><p>Smart has been out since March 21, depriving the Lakers of a key defensive player and a veteran leader. Redick said the Lakers had thought Smart would be back by now.</p><p>“He was doing on-court work, and he wanted to dial it back (last week),” Redick said. “He had a great day today on the court, trending towards playing at some point this week.”</p><p>The Lakers’ visit from the defending champion Thunder is the first of three games in four days. They play at Golden State on Thursday and host Phoenix on Friday.</p><p class="publish-time" id="end-of-article">Published on Apr 08, 2026</p></div> #Lakers #Oklahoma #City #LeBron #James #playing #NBA #today

Reddit’s legacy experience may be under threat.

The internet’s home page is constantly undergoing little changes here and there, but as Engadget reported, there’s reason to believe Old Reddit might not be long for this world in its current state.

The legacy version of the message board website, which some users prefer to the modern interface, will see unspecified “changes” over time, according to an official blog post. The idea is to combat AI scraping, which has been rampant on Reddit over the last few years.

Reddit has already made some changes to Old Reddit in this regard, as you currently can’t use Old Reddit without being logged into a real account.

One Reddit admin admitted recently that the company “can’t promise it will be around forever.” However, that doesn’t necessarily mean a total shutdown; CEO Steve Huffman also recently said Reddit is looking into limiting access or rebuilding it with modern tech rather than getting rid of it wholesale.

So, not only is AI making everything more expensive, but it’s also making it harder for older websites to exist. At a certain point, one has to wonder if it’s all worth it.

#Reddit #chopping #block #Heres">Old Reddit might be on the chopping block. Here’s why.
                                                            Reddit’s legacy experience may be under threat.The internet’s home page is constantly undergoing little changes here and there, but as Engadget reported, there’s reason to believe Old Reddit might not be long for this world in its current state. The legacy version of the message board website, which some users prefer to the modern interface, will see unspecified “changes” over time, according to an official blog post. The idea is to combat AI scraping, which has been rampant on Reddit over the last few years.
        
            Mashable Light Speed
        
        
    


        SEE ALSO:
        
            Reddit is ramping up its AI moderation tools. Moderators aren’t happy.
            
        
    
Reddit has already made some changes to Old Reddit in this regard, as you currently can’t use Old Reddit without being logged into a real account. One Reddit admin admitted recently that the company “can’t promise it will be around forever.” However, that doesn’t necessarily mean a total shutdown; CEO Steve Huffman also recently said Reddit is looking into limiting access or rebuilding it with modern tech rather than getting rid of it wholesale. 
So, not only is AI making everything more expensive, but it’s also making it harder for older websites to exist. At a certain point, one has to wonder if it’s all worth it.

                    
                                            
                            
    
        Topics
                    Artificial Intelligence
                    Reddit
            

                        
                                    #Reddit #chopping #block #Heres

Reddit’s legacy experience may be under threat.

The internet’s home page is constantly undergoing little changes here and there, but as Engadget reported, there’s reason to believe Old Reddit might not be long for this world in its current state.

The legacy version of the message board website, which some users prefer to the modern interface, will see unspecified “changes” over time, according to an official blog post. The idea is to combat AI scraping, which has been rampant on Reddit over the last few years.

Reddit has already made some changes to Old Reddit in this regard, as you currently can’t use Old Reddit without being logged into a real account.

One Reddit admin admitted recently that the company “can’t promise it will be around forever.” However, that doesn’t necessarily mean a total shutdown; CEO Steve Huffman also recently said Reddit is looking into limiting access or rebuilding it with modern tech rather than getting rid of it wholesale.

So, not only is AI making everything more expensive, but it’s also making it harder for older websites to exist. At a certain point, one has to wonder if it’s all worth it.

#Reddit #chopping #block #Heres">Old Reddit might be on the chopping block. Here’s why.

Reddit’s legacy experience may be under threat.

The internet’s home page is constantly undergoing little changes here and there, but as Engadget reported, there’s reason to believe Old Reddit might not be long for this world in its current state.

The legacy version of the message board website, which some users prefer to the modern interface, will see unspecified “changes” over time, according to an official blog post. The idea is to combat AI scraping, which has been rampant on Reddit over the last few years.

Reddit has already made some changes to Old Reddit in this regard, as you currently can’t use Old Reddit without being logged into a real account.

One Reddit admin admitted recently that the company “can’t promise it will be around forever.” However, that doesn’t necessarily mean a total shutdown; CEO Steve Huffman also recently said Reddit is looking into limiting access or rebuilding it with modern tech rather than getting rid of it wholesale.

So, not only is AI making everything more expensive, but it’s also making it harder for older websites to exist. At a certain point, one has to wonder if it’s all worth it.

#Reddit #chopping #block #Heres

Post Comment