ME

Melusine 103b i1

MelusineGGUF by mradermacherUpdated 2024-05-06
Download on Hugging Face
Parameters103B
Context32K
Layers120
Architecturellama
Smallest quant22.0 GB
Quick answer

Melusine 103b i1 needs about 39.6 GB of memory at IQ2_M (a 35.0 GB file) with an 8K context window, so it fits on a 48 GB GPU. This repository has 12 quants, from 22.0 GB (IQ1_S) to 46.4 GB (IQ3_M).

Melusine 103b i1 – a 103‑billion‑parameter Llama model in GGUF format

Melusine 103b i1 is a 103‑billion‑parameter Llama architecture model. It is quantized by mradermacher and available in multiple GGUF quantizations. The model supports roleplay tasks and is trained on a 32764‑token context.

Good at

  • Roleplay
  • Large context handling
  • High‑parameter reasoning

Limits

  • No vision support
  • License unspecified
  • Large memory requirement

Best for

  • Chatbots needing deep context
  • Roleplay applications
  • Complex reasoning tasks

Who it’s for. Developers and researchers who need a large Llama model in GGUF format for local inference. Suitable for users with sufficient GPU memory to run the chosen quantization.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_S22.0 GB 26.5 GB
Too big Get file →
IQ1_M24.1 GB 28.7 GB
Too big Get file →
IQ2_XXS27.6 GB 32.2 GB
Too big Get file →
IQ2_XS30.7 GB 35.2 GB
Too big Get file →
IQ2_S32.2 GB 36.8 GB
Too big Get file →
IQ2_M35.0 GB 39.6 GB
Too big Get file →
Q2_K38.2 GB 42.8 GB
Too big Get file →
IQ3_XXS39.9 GB 44.5 GB
Too big Get file →
IQ3_XS42.5 GB 47.0 GB
Too big Get file →
Q3_K_S44.8 GB 49.3 GB
Too big Get file →
IQ3_S44.9 GB 49.4 GB
Too big Get file →
IQ3_M46.4 GB 51.0 GB
Too big Get file →

For 8 GB VRAM, use IQ1_S (22.01 GB file, ~26.5 GB memory). For 12 GB VRAM, IQ1_M (24.13 GB file, ~28.7 GB memory). For 16 GB VRAM, IQ2_XXS (27.64 GB file, ~32.2 GB memory).

What GPU do you need for Melusine 103b i1?

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: the best quant that fits is IQ1_M (24.1 GB file, about 28.7 GB in use).
  • 48 GB VRAM: the best quant that fits is IQ3_XS (42.5 GB file, about 47.0 GB in use).

Which Melusine 103b i1 quant should you choose?

IQ2_M (35.0 GB) is the middle option in this repository. 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.

If memory is tight, the smallest file is IQ1_S at 22.0 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

Architecturellama
Parameters103B
Layers120
Trained context32,764 tokens
KV cache (fp16)480 KB per token

Source

Quantized bymradermacher
FamilyMelusine
Downloads (HF)1,335
Last updated2024-05-06

Frequently asked questions about Melusine 103b i1

How much VRAM does Melusine 103b i1 need?

About 39.6 GB at IQ2_M with an 8K context window. That is the 35.0 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 Melusine 103b i1 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 26.5 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Melusine 103b i1 GGUF?

It depends on the quant: IQ1_S is 22.0 GB, IQ1_M is 24.1 GB, IQ2_XXS is 27.6 GB, IQ2_XS is 30.7 GB, IQ2_S is 32.2 GB, IQ2_M is 35.0 GB and 6 more.

What context length does Melusine 103b i1 support?

The GGUF header lists a trained context length of 32,764 tokens (32K). The KV cache grows with the context you use: about 4.0 GB at 8K and 16.1 GB at 32K.

Where can I download Melusine 103b i1 GGUF?

The files are on Hugging Face in the mradermacher/Melusine_103b-i1-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

Can I run Melusine 103b i1 on 8 GB VRAM?

Use the IQ1_S quantization; it requires about 26.5 GB of VRAM at 8k context, so 8 GB is insufficient.

Does this model support vision tasks?

No, the model has no vision capabilities.

What is the context length of this model?

It was trained on 32764 tokens and can handle that context length.

Other GGUFs of Melusine_103b

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Melusine models

All Melusine →

Other large models

All large models →