arcee-ai / Trinity-Nano-Preview-MLX-6bit

huggingface.co
Total runs: 92
24-hour runs: 0
7-day runs: 33
30-day runs: 33
Model's Last Updated: May 29 2026
text-generation

Introduction of Trinity-Nano-Preview-MLX-6bit

Model Details of Trinity-Nano-Preview-MLX-6bit

Arcee Trinity Mini

Trinity Nano MLX 6bit

Trinity Nano Preview is a preview of Arcee AI's 6B MoE model with 1B active parameters. It is the small-sized model in our new Trinity family, a series of open-weight models for enterprise and tinkerers alike.

This is a chat tuned model, with a delightful personality and charm we think users will love. We note that this model is pushing the limits of sparsity in small language models with only 800M non-embedding parameters active per token, and as such may be unstable in certain use cases, especially in this preview.

This is an experimental release, it's fun to talk to but will not be hosted anywhere, so download it and try it out yourself!


Trinity Nano Preview is trained on 10T tokens gathered and curated through a key partnership with Datology , building upon the excellent dataset we used on AFM-4.5B with additional math and code.

Training was performed on a cluster of 512 H200 GPUs powered by Prime Intellect using HSDP parallelism.

More details, including key architecture decisions, can be found on our blog here


Model Details
  • Model Architecture: AfmoeForCausalLM
  • Parameters: 6B, 1B active
  • Experts: 128 total, 8 active, 1 shared
  • Context length: 128k
  • Training Tokens: 10T
  • License: Apache 2.0
Use with mlx
pip install mlx-lm
from mlx_lm import load, generate
from mlx_lm.sample_utils import make_sampler, make_logits_processors

model, tokenizer = load("arcee-ai/Trinity-Nano-Preview-MLX-6bit")

prompt = "What is the capital of France?"

if tokenizer.chat_template is not None:
    messages = [{"role": "user", "content": prompt}]
    prompt = tokenizer.apply_chat_template(
        messages, tokenize=False, add_generation_prompt=True
    )

sampler = make_sampler(temp=0.1, top_k=50, top_p=0.1)
logits_processors = make_logits_processors(repetition_penalty=1.05)

response = generate(
    model,
    tokenizer,
    prompt=prompt,
    max_tokens=512,
    sampler=sampler,
    logits_processors=logits_processors,
    verbose=True,
)

Runs of arcee-ai Trinity-Nano-Preview-MLX-6bit on huggingface.co

92
Total runs
0
24-hour runs
0
3-day runs
33
7-day runs
33
30-day runs

More Information About Trinity-Nano-Preview-MLX-6bit huggingface.co Model

More Trinity-Nano-Preview-MLX-6bit license Visit here:

https://choosealicense.com/licenses/openmdw-1.1

Trinity-Nano-Preview-MLX-6bit huggingface.co

Trinity-Nano-Preview-MLX-6bit huggingface.co is an AI model on huggingface.co that provides Trinity-Nano-Preview-MLX-6bit's model effect (), which can be used instantly with this arcee-ai Trinity-Nano-Preview-MLX-6bit model. huggingface.co supports a free trial of the Trinity-Nano-Preview-MLX-6bit model, and also provides paid use of the Trinity-Nano-Preview-MLX-6bit. Support call Trinity-Nano-Preview-MLX-6bit model through api, including Node.js, Python, http.

Trinity-Nano-Preview-MLX-6bit huggingface.co Url

https://huggingface.co/arcee-ai/Trinity-Nano-Preview-MLX-6bit

arcee-ai Trinity-Nano-Preview-MLX-6bit online free

Trinity-Nano-Preview-MLX-6bit huggingface.co is an online trial and call api platform, which integrates Trinity-Nano-Preview-MLX-6bit's modeling effects, including api services, and provides a free online trial of Trinity-Nano-Preview-MLX-6bit, you can try Trinity-Nano-Preview-MLX-6bit online for free by clicking the link below.

arcee-ai Trinity-Nano-Preview-MLX-6bit online free url in huggingface.co:

https://huggingface.co/arcee-ai/Trinity-Nano-Preview-MLX-6bit

Trinity-Nano-Preview-MLX-6bit install

Trinity-Nano-Preview-MLX-6bit is an open source model from GitHub that offers a free installation service, and any user can find Trinity-Nano-Preview-MLX-6bit on GitHub to install. At the same time, huggingface.co provides the effect of Trinity-Nano-Preview-MLX-6bit install, users can directly use Trinity-Nano-Preview-MLX-6bit installed effect in huggingface.co for debugging and trial. It also supports api for free installation.

Trinity-Nano-Preview-MLX-6bit install url in huggingface.co:

https://huggingface.co/arcee-ai/Trinity-Nano-Preview-MLX-6bit

Url of Trinity-Nano-Preview-MLX-6bit

Trinity-Nano-Preview-MLX-6bit huggingface.co Url

Provider of Trinity-Nano-Preview-MLX-6bit huggingface.co

arcee-ai
ORGANIZATIONS

Other API from arcee-ai

huggingface.co

Total runs: 18.6K
Run Growth: -4.8K
Growth Rate: -25.97%
Updated:May 29 2026
huggingface.co

Total runs: 7.7K
Run Growth: -3.4K
Growth Rate: -43.29%
Updated:September 18 2025
huggingface.co

Total runs: 3.1K
Run Growth: -968
Growth Rate: -30.44%
Updated:September 18 2025
huggingface.co

Total runs: 700
Run Growth: 280
Growth Rate: 40.00%
Updated:July 22 2024
huggingface.co

Total runs: 499
Run Growth: -150
Growth Rate: -30.06%
Updated:October 10 2025
huggingface.co

Total runs: 250
Run Growth: 27
Growth Rate: 10.71%
Updated:June 03 2025
huggingface.co

Total runs: 185
Run Growth: 28
Growth Rate: 15.14%
Updated:February 27 2025
huggingface.co

Total runs: 181
Run Growth: 135
Growth Rate: 74.59%
Updated:June 11 2025
huggingface.co

Total runs: 168
Run Growth: -15
Growth Rate: -8.93%
Updated:July 19 2024
huggingface.co

Total runs: 157
Run Growth: -18
Growth Rate: -11.46%
Updated:August 01 2024
huggingface.co

Total runs: 142
Run Growth: 9
Growth Rate: 6.34%
Updated:September 10 2024
huggingface.co

Total runs: 137
Run Growth: 14
Growth Rate: 10.22%
Updated:January 16 2026