Arcee Trinity Large
Hacker ApprovedRanked 27 of 39 open-weight language models by recommenders
Praised as a rare fully open-weights American model with strong value; critics say it is undertrained and weak on agentic tasks.
Praised as a rare fully open-weights American model with strong value; critics say it is undertrained and weak on agentic tasks.
21 people gave an opinion
- recommend it
- 16
- criticize it
- 5
- approval share
- 76%
Recommended in 5 of 13 months
- November 2025: not recommended
- December 2025: recommended
- January 2026: recommended
- February 2026: recommended
- March 2026: not recommended
- April 2026: recommended
- May 2026: not recommended
- June 2026: not recommended
- July 2026: recommended
- August 2026: not recommended
- September 2026: not recommended
- October 2026: not recommended
- November 2026: not recommended (month in progress)
Why it carries the badge
- 16 people recommend itNeeds 10 or more, met.
- 3.2 recommenders for every criticNeeds 2 to 1 or better, met.
- Recommended in 5 of 13 monthsNeeds 4 or more, met.
The same three rules apply to every product, and the badge can’t be bought. How the badge works
What people say about it
The reasons people give, with the number of comments that give each one. Open a reason to read what they said.
Why people recommend it
Each mark is one comment
Open-weight model from the US
“So refreshing to see open source models like this come from the US”
Comment from 28 Jan 2026 in “Trinity Large: American Open Source Foundation Model” “one of the first high-performing fully open weight American models”
Comment from 2 Apr 2026 in “Trinity Large Thinking” “Trinity Large is American and quite decent”
Comment from 15 Jul 2026 in “Inkling: Our Open-Weights Model” Low cost for its performance
“done something with $20m that is so good people think it must have been a $100m”
Comment from 29 Jan 2026 in “Trinity Large: American Open Source Foundation Model” “That's crazy affordable. Promising!”
Comment from 2 Apr 2026 in “Trinity Large Thinking” Promising, impressive results overall
“the model is coming close to QWEN and Deepseek. Pretty impressive”
Comment from 28 Jan 2026 in “Trinity Large: American Open Source Foundation Model” “This is a wonderful release.”
Comment from 28 Jan 2026 in “Trinity Large: American Open Source Foundation Model”
What people warn about
Each mark is one comment
Undertrained and not fully post-trained
“It's just undertrained and undersized (in terms of active parameters)”
Comment from 29 Jan 2026 in “Trinity Large: American Open Source Foundation Model” “Its still not fully post trained and its a non-reasoning model, but its worth keeping an eye on”
Comment from 22 Feb 2026 in “Ggml.ai joins Hugging Face to ensure the long-term progress of Local AI” “The models need another couple point releases worth of post-training to make useful agents”
Comment from 15 Jul 2026 in “Inkling: Our Open-Weights Model” Weak at agentic tasks
“The models need another couple point releases worth of post-training to make useful agents”
Comment from 15 Jul 2026 in “Inkling: Our Open-Weights Model” “it was pretty dumb: it failed to understand the order of jobs in a Github Action”
Comment from 29 Jan 2026 in “Trinity Large: American Open Source Foundation Model” “fairly mediocre score for a large model”
Comment from 2 Apr 2026 in “Trinity Large Thinking” Trails newer rival models
“but GLM's at 5.1 now. I hope they catch up.”
Comment from 2 Apr 2026 in “Arcee-AI Trinity-Large-Thinking” “fairly mediocre score for a large model”
Comment from 2 Apr 2026 in “Trinity Large Thinking”
Other open-weight language models people recommend
The four most recommended other open-weight language models, with the same counts for each, so you can compare like for like. The tick on each bar marks the 2 to 1 line the badge needs.
| Product | Recommend | Criticize | Approval | Months | Owned | Badge |
|---|---|---|---|---|---|---|
| Qwen 3.8 27B | 1,173 | 426 | 73% | 13 | 2 mo32 owners | Approved |
| Z.ai GLM | 1,043 | 384 | 73% | 13 | 2 mo52 owners | Approved |
| Kimi | 931 | 359 | 72% | 13 | 1 mo26 owners | Approved |
| Gemma 4 | 713 | 249 | 74% | 13 | 1 mo14 owners | Approved |
| Arcee Trinity LargeThis page | 16 | 5 | 76% | 5 | — | Approved |
What people said
23 of the 24 comments that mention it, newest first. Each links to the original comment.
“The models need another couple point releases worth of post-training to make useful agents”
“Trinity Large is American and quite decent”
“getting stuck in thinking loops with incredibly basic questions”
“or Arcee's Trinity Large?”
“Trinity Large Preview managed 17/25 on my agentic SQL benchmark”
“fairly mediocre score for a large model”
“but GLM's at 5.1 now. I hope they catch up.”
“one of the first high-performing fully open weight American models”
“That's crazy affordable. Promising!”
“Its still not fully post trained and its a non-reasoning model, but its worth keeping an eye on”
“done something with $20m that is so good people think it must have been a $100m”
“it was pretty dumb: it failed to understand the order of jobs in a Github Action”
“Accuracy and precision in math problems. So far so good.”
“I highly recommend. As a tip, you can quite easily get into a chat like state by simply using in context learning.”
“Waste of money towards ultimate goal: global loss of jobs and destroying earth.”
“It's just undertrained and undersized (in terms of active parameters)”
“With such a high throughput because of sparsity”
“some other optimizations that overall worked out really well”
“So refreshing to see open source models like this come from the US”
“This is a wonderful release.”
“the model is coming close to QWEN and Deepseek. Pretty impressive”
“I'm particularly excited to see a "true base" model to do research off of”
“Just perfect for a large Ram pool @ q4.”
Counts come from 1,795,640 Hacker News comments, November 2025 to November 2026. How we count. Report a mistake.