Sailor2 20B Chat needs about 14.3 GB of memory at Q4_K_M (a 11.6 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 11 quants, from 7.4 GB (Q2_K) to 20.4 GB (Q8_0).
Sailor2 20B Chat is a 19.2B-parameter Sailor model in GGUF format, quantized by mradermacher. Its trained context length is 4K tokens. The repository has 11 quant files, from 7.4 GB (Q2_K) to 20.4 GB (Q8_0).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 7.4 GB | 9.0 GB | Too big | Get file → | |
| Q3_K_S | 8.6 GB | 10.1 GB | Too big | Get file → | |
| Q3_K_M | 9.5 GB | 11.0 GB | Too big | Get file → | |
| Q3_K_L | 10.2 GB | 11.8 GB | Too big | Get file → | |
| IQ4_XS | 10.6 GB | 12.1 GB | Too big | Get file → | |
| Q4_K_S | 11.1 GB | 12.6 GB | Too big | Get file → | |
| Q4_K_M | 11.6 GB | 13.2 GB | Too big | Get file → | |
| Q5_K_S | 13.3 GB | 14.9 GB | Too big | Get file → | |
| Q5_K_M | 13.6 GB | 15.2 GB | Too big | Get file → | |
| Q6_K | 15.7 GB | 17.3 GB | Too big | Get file → | |
| Q8_0 | 20.4 GB | 22.0 GB | Too big | Get file → |
What GPU do you need for Sailor2 20B Chat?
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 Q3_K_S (8.6 GB file, about 11.2 GB in use).
- 16 GB VRAM: the best quant that fits is Q5_K_S (13.3 GB file, about 15.9 GB in use).
- 24 GB VRAM: the best quant that fits is Q8_0 (20.4 GB file, about 23.0 GB in use).
Which Sailor2 20B Chat quant should you choose?
Q4_K_M is the usual starting point for Sailor2 20B Chat: it balances file size and quality well. Here it is a 11.6 GB file.
The Q8_0 file is 20.4 GB, about 1.8 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 7.4 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
Source
Frequently asked questions about Sailor2 20B Chat
How much VRAM does Sailor2 20B Chat need?
About 14.3 GB at Q4_K_M with an 8K context window. That is the 11.6 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 Sailor2 20B Chat on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 10.0 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Sailor2 20B Chat GGUF?
It depends on the quant: Q2_K is 7.4 GB, Q3_K_S is 8.6 GB, Q3_K_M is 9.5 GB, Q3_K_L is 10.2 GB, IQ4_XS is 10.6 GB, Q4_K_S is 11.1 GB and 5 more.
What context length does Sailor2 20B Chat support?
The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 2.1 GB at 8K and 1.1 GB at 4K.
Where can I download Sailor2 20B Chat GGUF?
The files are on Hugging Face in the mradermacher/Sailor2-20B-Chat-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 Sailor2 20B Chat 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.