LL

Taigi Llama 2 Translator 7B

LlamaGGUF by RichardErkhovUpdated 2024-10-10
Download on Hugging Face
Parameters6.94B
Context4K
Layers32
Architecturellama
Smallest quant2.6 GB
Quick answer

Taigi Llama 2 Translator 7B needs about 9.0 GB of memory at Q4_K_M (a 4.2 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 22 quants, from 2.6 GB (Q2_K) to 7.4 GB (Q8_0).

Taigi Llama 2 Translator 7B – a 7B parameter translation model for Taiwanese Hokkien

Taigi Llama 2 Translator 7B is a 6.94B parameter model fine‑tuned on 263k parallel sentences. It translates between Traditional Chinese, English, and Taiwanese Hokkien in Hanzi, POJ, or Hanlo scripts. The model supports a 4096 token context window.

Good at

  • Translating Taiwanese Hokkien scripts
  • Bilingual translation between Chinese and English
  • Fine‑tuned for Taiwanese Hokkien

Limits

  • No vision support
  • License cc‑by‑nc‑sa‑4.0 limits commercial use
  • Requires 8k context for full performance

Best for

  • Language translation apps
  • Educational tools for Taiwanese Hokkien
  • Multilingual content creation

Who it’s for. Developers building translation services for Taiwanese Hokkien. Researchers needing a fine‑tuned Hokkien model for NLP tasks. Language enthusiasts creating bilingual content.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K2.6 GB 5.3 GB
Fits Get file →
IQ3_XS2.9 GB 5.6 GB
Fits Get file →
IQ3_S3.1 GB 5.7 GB
Fits Get file →
Q3_K_S3.1 GB 5.7 GB
Fits Get file →
IQ3_M3.2 GB 5.9 GB
Fits Get file →
Q3_K3.4 GB 6.1 GB
Fits Get file →
Q3_K_M3.4 GB 6.1 GB
Fits Get file →
Q3_K_L3.7 GB 6.4 GB
Fits Get file →
IQ4_XS3.8 GB 6.4 GB
Fits Get file →
Q4_04.0 GB 6.6 GB
Fits Get file →
IQ4_NL4.0 GB 6.6 GB
Fits Get file →
Q4_K_S4.0 GB 6.6 GB
Fits Get file →
Q4_K4.2 GB 6.9 GB
Fits Get file →
Q4_K_M4.2 GB 6.9 GB
Fits Get file →
Q4_14.4 GB 7.0 GB
Fits Get file →
Q5_04.8 GB 7.4 GB
Tight Get file →
Q5_K_S4.8 GB 7.4 GB
Tight Get file →
Q5_K4.9 GB 7.6 GB
Tight Get file →
Q5_K_M4.9 GB 7.6 GB
Tight Get file →
Q5_15.2 GB 7.9 GB
Tight Get file →
Q6_K5.7 GB 8.3 GB
Too big Get file →
Q8_07.4 GB 10.0 GB
Too big Get file →

For 8 GB VRAM, use Q4_0 (file size 3.96 GB, estimated 8.8 GB memory). For 12 GB VRAM, Q6_K (5.69 GB, 10.5 GB memory) or Q8_0 (7.37 GB, 12.2 GB memory). Smaller quants like Q2_K (2.65 GB, 7.4 GB memory) fit 8 GB VRAM but use more memory at 8k context.

What GPU do you need for Taigi Llama 2 Translator 7B?

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: the best quant that fits is IQ3_M (3.2 GB file, about 8.0 GB in use).
  • 12 GB VRAM: the best quant that fits is Q6_K (5.7 GB file, about 10.5 GB in use).
  • 16 GB VRAM: the best quant that fits is Q8_0 (7.4 GB file, about 12.2 GB in use).

Which Taigi Llama 2 Translator 7B quant should you choose?

Q4_K_M is the usual starting point for Taigi Llama 2 Translator 7B: it balances file size and quality well. Here it is a 4.2 GB file.

The Q8_0 file is 7.4 GB, about 1.7 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 Q2_K at 2.6 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
Parameters6.94B
Layers32
Trained context4,096 tokens
KV cache (fp16)512 KB per token

Source

Quantized byRichardErkhov
FamilyLlama
Downloads (HF)2,062
Last updated2024-10-10

Frequently asked questions about Taigi Llama 2 Translator 7B

How much VRAM does Taigi Llama 2 Translator 7B need?

About 9.0 GB at Q4_K_M with an 8K context window. That is the 4.2 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 Taigi Llama 2 Translator 7B on an 8 GB GPU?

Yes, with IQ3_M: the 3.2 GB file needs about 8.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Taigi Llama 2 Translator 7B GGUF?

It depends on the quant: Q2_K is 2.6 GB, IQ3_XS is 2.9 GB, IQ3_S is 3.1 GB, Q3_K_S is 3.1 GB, IQ3_M is 3.2 GB, Q3_K is 3.4 GB and 16 more.

What context length does Taigi Llama 2 Translator 7B support?

The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 4.3 GB at 8K and 2.1 GB at 4K.

Where can I download Taigi Llama 2 Translator 7B GGUF?

The files are on Hugging Face in the RichardErkhov/Bohanlu_-_Taigi-Llama-2-Translator-7B-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 Taigi Llama 2 Translator 7B use?

Hugging Face lists the license as cc-by-nc-sa-4.0. 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 this model on 8 GB VRAM?

Yes, using Q4_0 or smaller quants will fit within 8 GB VRAM, though memory usage is approximate.

Does this model support image input?

No, the model has no vision capabilities.

Is the model free to use?

It is licensed cc‑by‑nc‑sa‑4.0, allowing non‑commercial use with attribution.

Similar models

Browse all →

More from RichardErkhov

All by RichardErkhov →

More Llama models

All Llama →

Other small models

All small models →