RH

Rhea 72b v0.5

RheaGGUF by mradermacherUpdated 2024-05-06
Download on Hugging Face
Parameters72.3B
Context32K
Layers80
Architecturellama
Smallest quant31.0 GB
Quick answer

Rhea 72b v0.5 needs about 69.7 GB of memory at Q4_K_M (a 47.7 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 12 quants, from 31.0 GB (Q2_K) to 47.7 GB (Q4_K_M). Longer context needs more: at 32K tokens the KV cache alone adds about 85.9 GB.

Rhea 72b v0.5 – 72.3B parameter Llama model in GGUF format

Rhea 72b v0.5 is a 72.3B parameter Llama model fine‑tuned with SFT and DPO techniques. It supports a 32768 token context window and is distributed under the Apache‑2.0 license. The model is available in multiple quantizations for various memory budgets.

Good at

  • Large context generation
  • Fine‑tuned SFT and DPO performance
  • Open‑source licensing

Limits

  • No vision support
  • Requires high VRAM for larger quants
  • No mixture of experts

Best for

  • Research on long‑form text generation
  • Fine‑tuning with custom datasets
  • Benchmarking large‑scale LLMs

Who it’s for. Researchers and developers needing a high‑capacity Llama model for text‑only tasks. The model is suitable for those who can manage the memory requirements of large quantizations.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K31.0 GB 53.0 GB
Too big Get file →
IQ3_XS33.9 GB 55.9 GB
Too big Get file →
IQ3_S35.5 GB 57.5 GB
Too big Get file →
Q3_K_S35.5 GB 57.5 GB
Too big Get file →
IQ3_M37.2 GB 59.2 GB
Too big Get file →
Q3_K_M39.2 GB 61.2 GB
Too big Get file →
Q3_K_L42.5 GB 64.4 GB
Too big Get file →
IQ4_XS43.1 GB 65.0 GB
Too big Get file →
Q4_045.0 GB 66.9 GB
Too big Get file →
IQ4_NL45.2 GB 67.2 GB
Too big Get file →
Q4_K_S45.2 GB 67.2 GB
Too big Get file →
Q4_K_M47.7 GB 69.7 GB
Too big Get file →

For 8 GB VRAM, use Q2_K (31.04 GB file, 53 GB memory at 8k context). For 16 GB VRAM, Q3_K_S (35.53 GB file, 57.5 GB memory) or IQ3_S (35.53 GB file, 57.5 GB memory). For 24 GB VRAM, Q4_K_M (47.73 GB file, 69.7 GB memory). Memory estimates are approximate.

What GPU do you need for Rhea 72b 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: 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 Rhea 72b v0.5 quant should you choose?

Q4_K_M is the usual starting point for Rhea 72b v0.5: it balances file size and quality well. Here it is a 47.7 GB file.

If memory is tight, the smallest file is Q2_K at 31.0 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
Parameters72.3B
Layers80
Trained context32,768 tokens
KV cache (fp16)2560 KB per token

Source

Quantized bymradermacher
FamilyRhea
Licenseapache-2.0
Downloads (HF)577
Last updated2024-05-06

Frequently asked questions about Rhea 72b v0.5

How much VRAM does Rhea 72b v0.5 need?

About 69.7 GB at Q4_K_M with an 8K context window. That is the 47.7 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 Rhea 72b v0.5 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 53.0 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Rhea 72b v0.5 GGUF?

It depends on the quant: Q2_K is 31.0 GB, IQ3_XS is 33.9 GB, IQ3_S is 35.5 GB, Q3_K_S is 35.5 GB, IQ3_M is 37.2 GB, Q3_K_M is 39.2 GB and 6 more.

What context length does Rhea 72b v0.5 support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 21.5 GB at 8K and 85.9 GB at 32K.

Where can I download Rhea 72b v0.5 GGUF?

The files are on Hugging Face in the mradermacher/Rhea-72b-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 Rhea 72b v0.5 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 Rhea 72b v0.5 on 8 GB VRAM?

Only the Q2_K quant (31.04 GB file) can be loaded with 8 GB VRAM, but it still requires about 53 GB of memory at 8k context.

Does Rhea 72b v0.5 support vision tasks?

No, the model has no vision capabilities.

What license applies to Rhea 72b v0.5?

The model is released under the Apache‑2.0 license.

Other GGUFs of Rhea-72b-v0.5

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Rhea models

All Rhea →

Other large models

All large models →