SO

SoftTiger 13b

SofttigerGGUF by mradermacherUpdated 2024-12-10
Download on Hugging Face
Parameters13.4B
Context8K
Layers40
Architecturellama
Smallest quant3.1 GB
Quick answer

SoftTiger 13b needs about 15.3 GB of memory at Q4_K_M (a 8.1 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 25 quants, from 3.1 GB (IQ1_S) to 11.0 GB (Q6_K).

SoftTiger 13b – a 13.4 billion‑parameter LLaMA‑based model

SoftTiger 13b is a 13.4 billion‑parameter LLaMA architecture model trained with a 8192‑token context window. It is released under the apache‑2.0 license and available in multiple GGUF quantizations.

Good at

  • Large context handling
  • Flexible quantization options
  • Open‑source licensing

Limits

  • No vision support
  • Requires 8 k context for estimates
  • No mixture of experts

Best for

  • Chat applications
  • Text generation tasks
  • Fine‑tuning on custom data

Who it’s for. Developers and researchers who need a large LLaMA‑based model with a long context window and open‑source licensing. The model is suitable for local deployment via llama.cpp or similar frameworks.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_S3.1 GB 10.3 GB
Too big Get file →
IQ1_M3.3 GB 10.5 GB
Too big Get file →
IQ2_XXS3.7 GB 10.9 GB
Too big Get file →
IQ2_XS4.1 GB 11.3 GB
Too big Get file →
IQ2_S4.4 GB 11.6 GB
Too big Get file →
Q2_K_S4.6 GB 11.8 GB
Too big Get file →
IQ2_M4.7 GB 11.9 GB
Too big Get file →
Q2_K5.0 GB 12.3 GB
Too big Get file →
IQ3_XXS5.2 GB 12.4 GB
Too big Get file →
IQ3_XS5.6 GB 12.8 GB
Too big Get file →
IQ3_S5.9 GB 13.1 GB
Too big Get file →
Q3_K_S5.9 GB 13.1 GB
Too big Get file →
IQ3_M6.2 GB 13.4 GB
Too big Get file →
Q3_K_M6.6 GB 13.8 GB
Too big Get file →
Q3_K_L7.1 GB 14.4 GB
Too big Get file →
IQ4_XS7.2 GB 14.4 GB
Too big Get file →
Q4_07.6 GB 14.8 GB
Too big Get file →
Q4_07.6 GB 14.8 GB
Too big Get file →
Q4_07.6 GB 14.8 GB
Too big Get file →
Q4_07.6 GB 14.8 GB
Too big Get file →
Q4_K_S7.7 GB 14.9 GB
Too big Get file →
Q4_K_M8.1 GB 15.3 GB
Too big Get file →
Q5_K_S9.2 GB 16.4 GB
Too big Get file →
Q5_K_M9.5 GB 16.7 GB
Too big Get file →
Q6_K11.0 GB 18.2 GB
Too big Get file →

For 8 GB VRAM, use IQ1_S (3.07 GB file, 10.3 GB memory) or IQ1_M (3.31 GB file, 10.5 GB memory). If 12 GB VRAM is available, Q4_K_M (8.1 GB file, 15.3 GB memory) or Q5_K_M (9.49 GB file, 16.7 GB memory) are viable. Larger VRAM allows higher‑quality quants like Q6_K.

What GPU do you need for SoftTiger 13b?

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: the best quant that fits is IQ2_M (4.7 GB file, about 11.9 GB in use).
  • 16 GB VRAM: the best quant that fits is Q4_K_M (8.1 GB file, about 15.3 GB in use).
  • 24 GB VRAM: the best quant that fits is Q6_K (11.0 GB file, about 18.2 GB in use).

Which SoftTiger 13b quant should you choose?

Q4_K_M is the usual starting point for SoftTiger 13b: it balances file size and quality well. Here it is a 8.1 GB file.

If memory is tight, the smallest file is IQ1_S 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
Parameters13.4B
Layers40
Trained context8,192 tokens
KV cache (fp16)800 KB per token

Source

Quantized bymradermacher
FamilySofttiger
Licenseapache-2.0
Downloads (HF)694
Last updated2024-12-10

Frequently asked questions about SoftTiger 13b

How much VRAM does SoftTiger 13b need?

About 15.3 GB at Q4_K_M with an 8K context window. That is the 8.1 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 SoftTiger 13b on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 10.3 GB, so you would need a larger GPU or CPU offloading.

What is the file size of SoftTiger 13b GGUF?

It depends on the quant: IQ1_S is 3.1 GB, IQ1_M is 3.3 GB, IQ2_XXS is 3.7 GB, IQ2_XS is 4.1 GB, IQ2_S is 4.4 GB, Q2_K_S is 4.6 GB and 19 more.

What context length does SoftTiger 13b support?

The GGUF header lists a trained context length of 8,192 tokens (8K). The KV cache grows with the context you use: about 6.7 GB at 8K and 6.7 GB at 8K.

Where can I download SoftTiger 13b GGUF?

The files are on Hugging Face in the mradermacher/SoftTiger-13b-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 SoftTiger 13b 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 SoftTiger 13b on 8 GB VRAM?

Yes, using IQ1_S or IQ1_M quants will fit within 8 GB VRAM with estimated memory usage around 10.3–10.5 GB.

Does SoftTiger 13b support vision tasks?

No, the model does not include vision capabilities.

What license applies to SoftTiger 13b?

The model is released under the apache‑2.0 license.

Other GGUFs of SoftTiger-13b

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Softtiger models

All Softtiger →

Other medium models

All medium models →