Ling 3.0 Flash Base needs about 93.4 GB of memory at Q4_K_M (a 77.0 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 14 quants, from 0.5 GB to 90.5 GB (Q5_K_M). Longer context needs more: at 32K tokens the KV cache alone adds about 63.5 GB.
Ling 3.0 Flash Base is a 127B-parameter Ling model in GGUF format, quantized by mradermacher. Its trained context length is 256K tokens. The repository has 14 quant files, from 0.5 GB to 90.5 GB (Q5_K_M).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 0.5 GB | 16.9 GB | Too big | Get file → | |
| Q2_K | 46.5 GB | 62.9 GB | Too big | Get file → | |
| IQ3_S | 55.1 GB | 71.5 GB | Too big | Get file → | |
| Q3_K_S | 55.1 GB | 71.5 GB | Too big | Get file → | |
| IQ3_M | 55.7 GB | 72.0 GB | Too big | Get file → | |
| Q3_K_M | 60.8 GB | 77.1 GB | Too big | Get file → | |
| Q3_K_L | 66.0 GB | 82.4 GB | Too big | Get file → | |
| IQ4_XS | 68.0 GB | 84.4 GB | Too big | Get file → | |
| Q4_0 | 72.2 GB | 88.6 GB | Too big | Get file → | |
| Q4_K_S | 72.4 GB | 88.8 GB | Too big | Get file → | |
| Q4_K_M | 77.0 GB | 93.4 GB | Too big | Get file → | |
| Q4_1 | 80.0 GB | 96.3 GB | Too big | Get file → | |
| Q5_K_S | 87.9 GB | 104.3 GB | Too big | Get file → | |
| Q5_K_M | 90.5 GB | 106.8 GB | Too big | Get file → |
What GPU do you need for Ling 3.0 Flash Base?
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: the best quant that fits is the main file (0.5 GB file, about 16.9 GB in use).
- 32 GB VRAM: the best quant that fits is the main file (0.5 GB file, about 16.9 GB in use).
- 48 GB VRAM: the best quant that fits is the main file (0.5 GB file, about 16.9 GB in use).
Which Ling 3.0 Flash Base quant should you choose?
Q4_K_M is the usual starting point for Ling 3.0 Flash Base: it balances file size and quality well. Here it is a 77.0 GB file.
If memory is tight, the smallest file is the first one in the table at 0.5 GB, with the quality trade-offs that come with fewer bits.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Ling 3.0 Flash Base
How much VRAM does Ling 3.0 Flash Base need?
About 93.4 GB at Q4_K_M with an 8K context window. That is the 77.0 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 Ling 3.0 Flash Base on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 16.9 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Ling 3.0 Flash Base GGUF?
It depends on the quant: file is 0.5 GB, Q2_K is 46.5 GB, IQ3_S is 55.1 GB, Q3_K_S is 55.1 GB, IQ3_M is 55.7 GB, Q3_K_M is 60.8 GB and 8 more.
What context length does Ling 3.0 Flash Base support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 15.9 GB at 8K and 63.5 GB at 32K.
Where can I download Ling 3.0 Flash Base GGUF?
The files are on Hugging Face in the mradermacher/Ling-3.0-flash-base-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 Ling 3.0 Flash Base use?
Hugging Face lists the license as mit. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.