Check my GPU

LLMs ·

Trinity Mini (26B-A3B)

Parameters
26B (3.0B active)
Architecture
Mixture of experts
Max context
128K tokens
License
openmdw-1.1
Released
2025-12

How to run it

Quantized files · GGUF

Works in LM Studio, llama.cpp, Jan and Ollama. No sign-in needed.

huggingface.co/arcee-ai/Trinity-Mini-GGUF (opens in a new tab)

ollama run hf.co/arcee-ai/Trinity-Mini-GGUF

Original weights · Hugging Face

Full-precision files from Arcee AI, for fine-tuning, vLLM, Transformers.

huggingface.co/arcee-ai/Trinity-Mini (opens in a new tab)

No sign-in needed.

Links checked 5 days ago against Hugging Face, Ollama and GitHub.

Memory needed

Weights plus KV cache and runtime overhead. Leave a little extra free for the operating system and other apps.

KV cache: sliding-window hybrid: the KV estimate counts the global-attention layers plus a small allowance for the sliding-window layers.

Quantization Quality Weights Total at 8K context Total at 32K context Download
Q2_K Low 10.4 GB 11.2 GB 11.7 GB GGUF (opens in a new tab)
Q3_K_M Low 12.8 GB 13.6 GB 14.1 GB GGUF (opens in a new tab)
Q4_K_M Good 15.9 GB 16.7 GB 17.2 GB GGUF (opens in a new tab)
Q5_K_M High 18.5 GB 19.3 GB 19.8 GB GGUF (opens in a new tab)
Q6_K High 21.4 GB 22.2 GB 22.7 GB GGUF (opens in a new tab)
Q8_0 Very high 27.7 GB 28.5 GB 29.0 GB GGUF (opens in a new tab)
F16 Full 52.2 GB 53.0 GB 53.5 GB Original weights (opens in a new tab)

Best hardware for Trinity Mini (26B-A3B)

The least expensive GPU in our catalog that runs it without offloading is the Intel Arc B580 (launch price $249) at Q2_K.

Grade · points Hardware Quant Memory Speed Fit
84 ptsGreat (A) · 84 of 100 points: Model quality 29.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 Apple M4 Max (64 GB) Q8_0 32.3 GB / 48.0 GB 49.0 tok/s Runs well
84 ptsGreat (A) · 84 of 100 points: Model quality 29.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 NVIDIA GeForce RTX 5090 Q6_K 23.8 GB / 32.0 GB 254 tok/s Runs well
83 ptsGreat (A) · 83 of 100 points: Model quality 29.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 13.9/15 NVIDIA RTX PRO 6000 Blackwell F16 57.8 GB / 96.0 GB 86.0 tok/s Runs well
83 ptsGreat (A) · 83 of 100 points: Model quality 29.4/45, Memory fit 24.0/25, Speed 15.0/15, Memory use 15.0/15 AMD Radeon RX 7900 XTX Q6_K 23.4 GB / 24.0 GB 117 tok/s Tight fit
83 ptsGreat (A) · 83 of 100 points: Model quality 29.4/45, Memory fit 24.0/25, Speed 15.0/15, Memory use 15.0/15 NVIDIA GeForce RTX 3090 Q6_K 23.4 GB / 24.0 GB 133 tok/s Tight fit
83 ptsGreat (A) · 83 of 100 points: Model quality 29.4/45, Memory fit 24.0/25, Speed 15.0/15, Memory use 15.0/15 NVIDIA GeForce RTX 4090 Q6_K 23.4 GB / 24.0 GB 143 tok/s Tight fit
82 ptsGreat (A) · 82 of 100 points: Model quality 28.8/45, Memory fit 25.0/25, Speed 14.9/15, Memory use 14.2/15 Apple M4 Pro (48 GB) Q5_K_M 22.2 GB / 36.0 GB 39.7 tok/s Runs well
81 ptsGreat (A) · 81 of 100 points: Model quality 29.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 12.5/15 Apple M1 Max (64 GB) Q6_K 26.0 GB / 48.0 GB 48.9 tok/s Runs well
78 ptsGreat (A) · 78 of 100 points: Model quality 24.9/45, Memory fit 24.0/25, Speed 15.0/15, Memory use 15.0/15 AMD Radeon RX 9070 XT Q3_K_M 14.4 GB / 16.0 GB 146 tok/s Tight fit
78 ptsGreat (A) · 78 of 100 points: Model quality 24.9/45, Memory fit 24.0/25, Speed 15.0/15, Memory use 15.0/15 Apple M4 Pro (24 GB) Q3_K_M 14.9 GB / 16.1 GB 62.2 tok/s Tight fit

Check it on your own hardware

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.