DU

Duet Minitron8b v0.5

DuetGGUF by bartowskiUpdated 2024-08-30
Download on Hugging Face
Parameters8.41B
Context8K
Layers40
Architecturellama
Smallest quant3.1 GB
Quick answer

Duet Minitron8b v0.5 needs about 7.0 GB of memory at Q4_K_M (a 5.1 GB file) with an 8K context window, so it fits on a 8 GB GPU. This repository has 24 quants, from 3.1 GB (IQ2_M) to 16.8 GB (F16).

Duet Minitron8b v0.5 – 8.41B llama model in GGUF format

Duet Minitron8b v0.5 is an 8.41‑billion‑parameter llama model quantized for llama.cpp. It supports a 8192‑token context window and is released under cc‑by‑2.0. The model is available in 24 quantization variants.

Good at

  • Large context reasoning
  • Multilingual text generation
  • Fine‑tuned prompt handling

Limits

  • No vision support
  • Requires 8‑k context for estimates
  • CC‑BY‑2.0 license

Best for

  • Chatbots with long context
  • Research prototyping
  • Educational demos

Who it’s for. Developers and researchers who need a large‑parameter llama model with flexible quantization options. The model is suitable for local inference on machines with sufficient VRAM and RAM.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ2_M3.1 GB 4.9 GB
Fits Get file →
Q2_K3.3 GB 5.2 GB
Fits Get file →
IQ3_XS3.7 GB 5.5 GB
Fits Get file →
Q3_K_S3.8 GB 5.7 GB
Fits Get file →
Q2_K_L3.9 GB 5.7 GB
Fits Get file →
IQ3_M4.0 GB 5.8 GB
Fits Get file →
Q3_K_M4.2 GB 6.1 GB
Fits Get file →
Q3_K_L4.5 GB 6.4 GB
Fits Get file →
IQ4_XS4.7 GB 6.5 GB
Fits Get file →
Q4_04.9 GB 6.7 GB
Fits Get file →
Q4_04.9 GB 6.7 GB
Fits Get file →
Q4_04.9 GB 6.7 GB
Fits Get file →
Q4_04.9 GB 6.7 GB
Fits Get file →
Q4_K_S4.9 GB 6.8 GB
Fits Get file →
Q3_K_XL5.0 GB 6.8 GB
Fits Get file →
Q4_K_M5.1 GB 7.0 GB
Fits Get file →
Q4_K_L5.5 GB 7.4 GB
Tight Get file →
Q5_K_S5.9 GB 7.7 GB
Tight Get file →
Q5_K_M6.0 GB 7.8 GB
Tight Get file →
Q5_K_L6.3 GB 8.2 GB
Too big Get file →
Q6_K6.9 GB 8.8 GB
Too big Get file →
Q6_K_L7.2 GB 9.0 GB
Too big Get file →
Q8_08.9 GB 10.8 GB
Too big Get file →
F1616.8 GB 18.7 GB
Too big Get file →

For 8‑GB VRAM, use Q4_K_M (file size 5.15 GB, memory 7 GB). For 12‑GB VRAM, Q5_K_M (6 GB, 7.8 GB) or Q6_K (6.91 GB, 8.8 GB). For 16‑GB VRAM, Q8_0 (8.95 GB, 10.8 GB) or F16 (16.84 GB, 18.7 GB). Estimates are approximate.

What GPU do you need for Duet Minitron8b v0.5?

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: the best quant that fits is IQ3_M (4.0 GB file, about 5.8 GB in use).
  • 8 GB VRAM: the best quant that fits is Q5_K_M (6.0 GB file, about 7.8 GB in use).
  • 12 GB VRAM: the best quant that fits is Q8_0 (8.9 GB file, about 10.8 GB in use).
  • 16 GB VRAM: the best quant that fits is Q8_0 (8.9 GB file, about 10.8 GB in use).
  • 24 GB VRAM: the best quant that fits is F16 (16.8 GB file, about 18.7 GB in use).

Which Duet Minitron8b v0.5 quant should you choose?

Q4_K_M is the usual starting point for Duet Minitron8b v0.5: it balances file size and quality well. Here it is a 5.1 GB file.

The Q8_0 file is 8.9 GB, about 1.7 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 IQ2_M at 3.1 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

Architecturellama
Parameters8.41B
Layers40
Trained context8,192 tokens
KV cache (fp16)160 KB per token

Source

Quantized bybartowski
FamilyDuet
Licensecc-by-2.0
Downloads (HF)609
Last updated2024-08-30

Frequently asked questions about Duet Minitron8b v0.5

How much VRAM does Duet Minitron8b v0.5 need?

About 7.0 GB at Q4_K_M with an 8K context window. That is the 5.1 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 Duet Minitron8b v0.5 on an 8 GB GPU?

Yes, with Q5_K_M: the 6.0 GB file needs about 7.8 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Duet Minitron8b v0.5 GGUF?

It depends on the quant: IQ2_M is 3.1 GB, Q2_K is 3.3 GB, IQ3_XS is 3.7 GB, Q3_K_S is 3.8 GB, Q2_K_L is 3.9 GB, IQ3_M is 4.0 GB and 18 more.

What context length does Duet Minitron8b v0.5 support?

The GGUF header lists a trained context length of 8,192 tokens (8K). The KV cache grows with the context you use: about 1.3 GB at 8K and 1.3 GB at 8K.

Where can I download Duet Minitron8b v0.5 GGUF?

The files are on Hugging Face in the bartowski/Duet_Minitron8b_v0.5-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 Duet Minitron8b v0.5 use?

Hugging Face lists the license as cc-by-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 Duet Minitron8b v0.5 on 8 GB VRAM?

Yes, the Q4_K_M quantization uses about 7 GB of VRAM at 8‑k context, fitting within an 8‑GB budget.

What license does the model use?

The model is released under cc‑by‑2.0, allowing reuse with attribution.

Does the model support vision tasks?

No, the model has no vision capabilities and is text‑only.

Other GGUFs of Duet_Minitron8b_v0.5

Similar models

Browse all →

More from bartowski

All by bartowski →

More Duet models

All Duet →

Other small models

All small models →