LL

Discoleo

LlamaGGUF by mradermacherUpdated 2024-09-11
Download on Hugging Face
Parameters8B
Context8K
Layers32
Architecturellama
Smallest quant2.0 GB
Quick answer

Discoleo needs about 6.5 GB of memory at Q4_K_M (a 4.9 GB file) with an 8K context window, so it fits on a 8 GB GPU. This repository has 24 quants, from 2.0 GB (IQ1_S) to 6.6 GB (Q6_K).

Discoleo – 8B instruction‑tuned Llama3 model

Discoleo is an 8B instruction‑tuned Llama3 model with 8192 context length. It is derived from Meta’s Llama3‑8B and fine‑tuned on German instruction data. The model is available in multiple GGUF quantizations.

Good at

  • German instruction following
  • Long‑context 8192 tokens
  • LLM chat templates

Limits

  • No vision support
  • License is llama3
  • Requires 8k context for full performance

Best for

  • German chatbots
  • Instruction‑driven applications
  • Fine‑tuned German LLM use cases

Who it’s for. Developers building German‑language LLM applications who need an instruction‑tuned model with a large context window. The model is suitable for research and production use where the llama3 license is acceptable.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_S2.0 GB 3.6 GB
Fits Get file →
IQ1_M2.2 GB 3.7 GB
Fits Get file →
IQ2_XXS2.4 GB 4.0 GB
Fits Get file →
IQ2_XS2.6 GB 4.2 GB
Fits Get file →
IQ2_S2.8 GB 4.3 GB
Fits Get file →
IQ2_M2.9 GB 4.5 GB
Fits Get file →
Q2_K3.2 GB 4.8 GB
Fits Get file →
IQ3_XXS3.3 GB 4.8 GB
Fits Get file →
IQ3_XS3.5 GB 5.1 GB
Fits Get file →
Q3_K_S3.7 GB 5.2 GB
Fits Get file →
IQ3_S3.7 GB 5.3 GB
Fits Get file →
IQ3_M3.8 GB 5.4 GB
Fits Get file →
Q3_K_M4.0 GB 5.6 GB
Fits Get file →
Q3_K_L4.3 GB 5.9 GB
Fits Get file →
IQ4_XS4.4 GB 6.0 GB
Fits Get file →
Q4_04.7 GB 6.2 GB
Fits Get file →
Q4_04.7 GB 6.2 GB
Fits Get file →
Q4_04.7 GB 6.2 GB
Fits Get file →
Q4_04.7 GB 6.2 GB
Fits Get file →
Q4_K_S4.7 GB 6.3 GB
Fits Get file →
Q4_K_M4.9 GB 6.5 GB
Fits Get file →
Q5_K_S5.6 GB 7.2 GB
Fits Get file →
Q5_K_M5.7 GB 7.3 GB
Tight Get file →
Q6_K6.6 GB 8.2 GB
Too big Get file →

For 8‑GB VRAM use IQ1_S (2.02 GB, 3.6 GB memory at 8k). For 12‑GB VRAM use Q4_K_M (4.92 GB, 6.5 GB memory). For 16‑GB VRAM use Q5_K_M (5.73 GB, 7.3 GB memory). Memory estimates are approximate.

What GPU do you need for Discoleo?

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: the best quant that fits is IQ2_XXS (2.4 GB file, about 4.0 GB in use).
  • 6 GB VRAM: the best quant that fits is IQ4_XS (4.4 GB file, about 6.0 GB in use).
  • 8 GB VRAM: the best quant that fits is Q5_K_M (5.7 GB file, about 7.3 GB in use).
  • 12 GB VRAM: the best quant that fits is Q6_K (6.6 GB file, about 8.2 GB in use).

Which Discoleo quant should you choose?

Q4_K_M is the usual starting point for Discoleo: it balances file size and quality well. Here it is a 4.9 GB file.

If memory is tight, the smallest file is IQ1_S at 2.0 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
Parameters8B
Layers32
Trained context8,192 tokens
KV cache (fp16)128 KB per token

Source

Quantized bymradermacher
FamilyLlama
Licensellama3
Downloads (HF)1,842
Last updated2024-09-11

Frequently asked questions about Discoleo

How much VRAM does Discoleo need?

About 6.5 GB at Q4_K_M with an 8K context window. That is the 4.9 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 Discoleo on an 8 GB GPU?

Yes, with Q5_K_M: the 5.7 GB file needs about 7.3 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Discoleo GGUF?

It depends on the quant: IQ1_S is 2.0 GB, IQ1_M is 2.2 GB, IQ2_XXS is 2.4 GB, IQ2_XS is 2.6 GB, IQ2_S is 2.8 GB, IQ2_M is 2.9 GB and 18 more.

What context length does Discoleo support?

The GGUF header lists a trained context length of 8,192 tokens (8K). The KV cache grows with the context you use: about 1.1 GB at 8K and 1.1 GB at 8K.

Where can I download Discoleo GGUF?

The files are on Hugging Face in the mradermacher/Llama3-DiscoLeo-Instruct-8B-v0.1-i1-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 Discoleo use?

Hugging Face lists the license as llama3. 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 Discoleo on 8 GB VRAM?

Yes, use IQ1_S or IQ1_M; they require about 3.6–3.7 GB memory at 8k context.

Does Discoleo support vision tasks?

No, the model has no vision capabilities.

What is the maximum context length?

The model supports up to 8192 tokens.

Other GGUFs of Llama3-DiscoLeo-Instruct-8B-v0.1

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Llama models

All Llama →

Other small models

All small models →