CA

calme-3.1-instruct-78b

CalmeGGUF by bartowskiUpdated 2024-11-21
Download on Hugging Face
Parameters78B
Context32K
Layers86
Architectureqwen2
Smallest quant25.4 GB
Quick answer

calme-3.1-instruct-78b needs about 54.1 GB of memory at Q4_K_M (a 50.7 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 21 quants, from 25.4 GB (IQ1_M) to 82.9 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 11.5 GB.

Calme 3.1 Instruct 78B – a large Qwen2 based instruction model

Calme-3.1-instruct-78b is a 78B parameter model based on the Qwen2 architecture. It is fine‑tuned from Qwen2.5‑72B and supports a 32,768 token context window. The model is available in multiple GGUF quantizations for llama.cpp.

Good at

  • Large context handling
  • Instruction following
  • General domain knowledge

Limits

  • No vision support
  • Experimental status may affect performance
  • License is other

Best for

  • Chatbot development
  • Content generation
  • Research prototyping

Who it’s for. Developers and researchers who need a large instruction‑tuned model for text‑only tasks and can manage the memory requirements. The model is suitable for those who prefer open‑source quantizations and are comfortable with experimental releases.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_M25.4 GB 28.8 GB
Too big Get file →
IQ2_XXS27.3 GB 30.7 GB
Too big Get file →
IQ2_XS29.0 GB 32.4 GB
Too big Get file →
IQ2_M31.4 GB 34.8 GB
Too big Get file →
Q2_K31.8 GB 35.2 GB
Too big Get file →
Q2_K_L33.1 GB 36.4 GB
Too big Get file →
IQ3_XXS34.0 GB 37.4 GB
Too big Get file →
Q3_K_S36.8 GB 40.2 GB
Too big Get file →
IQ3_M37.9 GB 41.3 GB
Too big Get file →
Q3_K_M40.3 GB 43.7 GB
Too big Get file →
Q3_K_L42.3 GB 45.7 GB
Too big Get file →
IQ4_XS42.6 GB 45.9 GB
Too big Get file →
Q3_K_XL43.4 GB 46.8 GB
Too big Get file →
Q4_044.2 GB 47.6 GB
Too big Get file →
Q4_044.3 GB 47.7 GB
Too big Get file →
Q4_K_S46.9 GB 50.3 GB
Too big Get file →
Q4_K_M50.7 GB (2 parts) 54.1 GB
Too big Get file →
Q5_K_S55.1 GB (2 parts) 58.5 GB
Too big Get file →
Q5_K_M58.3 GB (2 parts) 61.7 GB
Too big Get file →
Q6_K69.0 GB (2 parts) 72.4 GB
Too big Get file →
Q8_082.9 GB (3 parts) 86.2 GB
Too big Get file →

For 8 GB VRAM use Q4_0 (44.19 GB file, ~47.6 GB memory at 8 k context). For 16 GB VRAM Q5_K_S (55.08 GB file, ~58.5 GB memory) or Q5_K_M (58.31 GB file, ~61.7 GB memory). For 24 GB VRAM Q6_K (69.01 GB file, ~72.4 GB memory).

What GPU do you need for calme-3.1-instruct-78b?

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 IQ2_XXS (27.3 GB file, about 30.7 GB in use).
  • 48 GB VRAM: the best quant that fits is Q4_0 (44.3 GB file, about 47.7 GB in use).

Which calme-3.1-instruct-78b quant should you choose?

Q4_K_M is the usual starting point for calme-3.1-instruct-78b: it balances file size and quality well. Here it is a 50.7 GB file.

The Q8_0 file is 82.9 GB, about 1.6 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 IQ1_M at 25.4 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

Architectureqwen2
Parameters78B
Layers86
Trained context32,768 tokens
KV cache (fp16)344 KB per token

Source

Quantized bybartowski
FamilyCalme
Licenseother
Downloads (HF)904
Last updated2024-11-21

Frequently asked questions about calme-3.1-instruct-78b

How much VRAM does calme-3.1-instruct-78b need?

About 54.1 GB at Q4_K_M with an 8K context window. That is the 50.7 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 calme-3.1-instruct-78b on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 28.8 GB, so you would need a larger GPU or CPU offloading.

What is the file size of calme-3.1-instruct-78b GGUF?

It depends on the quant: IQ1_M is 25.4 GB, IQ2_XXS is 27.3 GB, IQ2_XS is 29.0 GB, IQ2_M is 31.4 GB, Q2_K is 31.8 GB, Q2_K_L is 33.1 GB and 15 more.

What context length does calme-3.1-instruct-78b support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 2.9 GB at 8K and 11.5 GB at 32K.

Where can I download calme-3.1-instruct-78b GGUF?

The files are on Hugging Face in the bartowski/calme-3.1-instruct-78b-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 calme-3.1-instruct-78b use?

Hugging Face lists the license as other. 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 this model on 8 GB VRAM?

Use the Q4_0 quantization; it requires about 47.6 GB memory at 8 k context, so 8 GB VRAM is insufficient.

Does this model support image inputs?

No, the model has no vision capabilities.

What is the context length of this model?

The model supports a 32,768 token context window.

Other GGUFs of calme-3.1-instruct-78b

Similar models

Browse all →

More from bartowski

All by bartowski →

More Calme models

All Calme →

Other large models

All large models →