TR

Trinity-Large-Preview

TrinityGGUF by unslothUpdated 2026-01-28
Download on Hugging Face
Parameters399B
Context256K
Layers60
Architectureafmoe
Smallest quant145.0 GB
Quick answer

Trinity-Large-Preview needs about 242.4 GB of memory at Q4_K_M (a 239.9 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 24 quants, from 145.0 GB (Q2_K) to 797.4 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 8.1 GB.

Trinity‑Large‑Preview: 398B‑parameter sparse MoE model with 13B active params per token

Trinity‑Large‑Preview is a 398B‑parameter sparse Mixture‑of‑Experts model. It was lightly post‑trained on 20B tokens and supports long‑context comprehension up to 262144 tokens. The model is chat‑ready and available under the Apache‑2.0 license.

Good at

  • Long‑context reasoning
  • Sparse MoE efficiency
  • Chat‑ready instruction tuning

Limits

  • Requires large VRAM for high‑quality quants
  • No vision support
  • Limited to 262144 token context

Best for

  • Chat applications needing long context
  • Research on MoE architectures
  • Large‑scale text generation

Who it’s for. Developers and researchers who need a large, sparse MoE model for long‑context tasks. The model is suitable for environments with sufficient GPU memory and for those who can manage the complexity of MoE inference.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K145.0 GB (3 parts) 147.5 GB
Too big Get file →
Q2_K_L145.1 GB (3 parts) 147.7 GB
Too big Get file →
Q2_K_XL145.8 GB (3 parts) 148.4 GB
Too big Get file →
IQ2_XXS153.5 GB (4 parts) 156.0 GB
Too big Get file →
IQ2_M154.3 GB (4 parts) 156.8 GB
Too big Get file →
Q3_K_S172.0 GB (4 parts) 174.5 GB
Too big Get file →
IQ3_XXS182.5 GB (4 parts) 185.1 GB
Too big Get file →
Q3_K_M189.4 GB (4 parts) 192.0 GB
Too big Get file →
Q3_K_XL189.9 GB (4 parts) 192.5 GB
Too big Get file →
IQ4_XS212.5 GB (5 parts) 215.0 GB
Too big Get file →
IQ4_NL224.8 GB (5 parts) 227.3 GB
Too big Get file →
Q4_0225.0 GB (5 parts) 227.5 GB
Too big Get file →
Q4_K_S225.1 GB (5 parts) 227.6 GB
Too big Get file →
Q4_K_M239.9 GB (5 parts) 242.4 GB
Too big Get file →
Q4_K_XL247.3 GB (5 parts) 249.8 GB
Too big Get file →
Q4_1249.6 GB (6 parts) 252.1 GB
Too big Get file →
Q5_K_S274.5 GB (6 parts) 277.0 GB
Too big Get file →
Q5_K_M282.3 GB (6 parts) 284.8 GB
Too big Get file →
Q5_K_XL284.8 GB (6 parts) 287.3 GB
Too big Get file →
Q6_K327.3 GB (7 parts) 329.8 GB
Too big Get file →
Q6_K_XL341.9 GB (7 parts) 344.4 GB
Too big Get file →
Q8_0423.7 GB (9 parts) 426.2 GB
Too big Get file →
Q8_K_XL467.3 GB (10 parts) 469.8 GB
Too big Get file →
BF16797.4 GB (17 parts) 799.9 GB
Too big Get file →

For 8‑GB VRAM, use Q2_K or Q2_K_L quants; they require about 147.5‑147.7 GB of memory at 8K context. For 16‑GB VRAM, Q3_K_S or Q3_K_M are suitable, needing roughly 174.5‑192 GB. For 32‑GB VRAM, Q4_K_S or Q4_K_M offer a balance between size and performance.

What GPU do you need for Trinity-Large-Preview?

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: no quant fits at an 8K context.
  • 24 GB VRAM: no quant fits at an 8K context.
  • 32 GB VRAM: no quant fits at an 8K context.
  • 48 GB VRAM: no quant fits at an 8K context.

Which Trinity-Large-Preview quant should you choose?

Q4_K_M is the usual starting point for Trinity-Large-Preview: it balances file size and quality well. Here it is a 239.9 GB file.

The Q8_0 file is 423.7 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 Q2_K at 145.0 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureafmoe
Parameters399B (approximate, mixture of experts)
Layers60
Trained context262,144 tokens
KV cache (fp16)240 KB per token

Source

Quantized byunsloth
FamilyTrinity
Licenseapache-2.0
Downloads (HF)1,184
Last updated2026-01-28

Frequently asked questions about Trinity-Large-Preview

How much VRAM does Trinity-Large-Preview need?

About 242.4 GB at Q4_K_M with an 8K context window. That is the 239.9 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 Trinity-Large-Preview on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 147.5 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Trinity-Large-Preview GGUF?

It depends on the quant: Q2_K is 145.0 GB, Q2_K_L is 145.1 GB, Q2_K_XL is 145.8 GB, IQ2_XXS is 153.5 GB, IQ2_M is 154.3 GB, Q3_K_S is 172.0 GB and 18 more.

What context length does Trinity-Large-Preview support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 2.0 GB at 8K and 8.1 GB at 32K.

Where can I download Trinity-Large-Preview GGUF?

The files are on Hugging Face in the unsloth/Trinity-Large-Preview-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 Trinity-Large-Preview 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.

Can I run Trinity‑Large‑Preview on 8 GB VRAM?

Only the lowest‑quality quants such as Q2_K or Q2_K_L can be loaded, but they still require roughly 147 GB of memory at 8K context, so 8 GB VRAM is insufficient.

Does the model support vision tasks?

No, Trinity‑Large‑Preview does not include vision capabilities.

What license applies to this model?

The model is released under the Apache‑2.0 license.

Other GGUFs of Trinity-Large-Preview

Similar models

Browse all →

More from unsloth

All by unsloth →

More Trinity models

All Trinity →

Other large models

All large models →