TR

Tri 21B Think

TriGGUF by mykorUpdated 2026-03-07
Download on Hugging Face
Parameters20.7B
Context32K
Layers40
Architecturellama
Smallest quant10.9 GB
Quick answer

Tri 21B Think needs about 14.8 GB of memory at Q4_K_M (a 12.6 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 6 quants, from 10.9 GB (Q3_K_L) to 41.5 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 6.7 GB.

Tri 21B Think is a 20.7B-parameter Tri model in GGUF format, quantized by mykor. Its trained context length is 32K tokens. The repository has 6 quant files, from 10.9 GB (Q3_K_L) to 41.5 GB (BF16).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q3_K_L10.9 GB 13.1 GB
Too big Get file →
Q4_K_M12.6 GB 14.8 GB
Too big Get file →
Q5_K_M14.7 GB 16.9 GB
Too big Get file →
Q6_K17.0 GB 19.2 GB
Too big Get file →
Q8_022.0 GB 24.2 GB
Too big Get file →
BF1641.5 GB 43.6 GB
Too big Get file →

What GPU do you need for Tri 21B Think?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: no quant fits at an 8K context.
  • 6 GB VRAM: no quant fits at an 8K context.
  • 8 GB VRAM: no quant fits at an 8K context.
  • 12 GB VRAM: no quant fits at an 8K context.
  • 16 GB VRAM: the best quant that fits is Q4_K_M (12.6 GB file, about 14.8 GB in use).
  • 24 GB VRAM: the best quant that fits is Q6_K (17.0 GB file, about 19.2 GB in use).
  • 32 GB VRAM: the best quant that fits is Q8_0 (22.0 GB file, about 24.2 GB in use).
  • 48 GB VRAM: the best quant that fits is BF16 (41.5 GB file, about 43.6 GB in use).

Which Tri 21B Think quant should you choose?

Q4_K_M is the usual starting point for Tri 21B Think: it balances file size and quality well. Here it is a 12.6 GB file.

The Q8_0 file is 22.0 GB, about 1.8 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is Q3_K_L at 10.9 GB, with the quality trade-offs that come with fewer bits. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters20.7B
Layers40
Trained context32,768 tokens
KV cache (fp16)200 KB per token

Source

Quantized bymykor
FamilyTri
Licenseapache-2.0
Downloads (HF)142
Last updated2026-03-07

Frequently asked questions about Tri 21B Think

How much VRAM does Tri 21B Think need?

About 14.8 GB at Q4_K_M with an 8K context window. That is the 12.6 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run Tri 21B Think on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 13.1 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Tri 21B Think GGUF?

It depends on the quant: Q3_K_L is 10.9 GB, Q4_K_M is 12.6 GB, Q5_K_M is 14.7 GB, Q6_K is 17.0 GB, Q8_0 is 22.0 GB, BF16 is 41.5 GB.

What context length does Tri 21B Think support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 1.7 GB at 8K and 6.7 GB at 32K.

Where can I download Tri 21B Think GGUF?

The files are on Hugging Face in the mykor/Tri-21B-Think-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does Tri 21B Think use?

Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Similar models

Browse all →

More from mykor

All by mykor →

More Tri models

All Tri →

Other medium models

All medium models →