Ling Flash 2.0 needs about 63.7 GB of memory at Q4_K_M (a 62.6 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 27 quants, from 0.3 GB to 109.4 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 2.1 GB.
Ling Flash 2.0 – 100B parameter MoE LLM with 6.1B active parameters
Ling Flash 2.0 is a 100B parameter mixture of experts model with a 1/32 activation ratio. It was trained on 20T+ tokens and achieves strong performance in complex reasoning, code generation, and frontend development. The model is released under the MIT license.
Good at
- Complex reasoning
- Code generation
- Frontend development
Limits
- Large VRAM requirement for full precision
- No vision support
- Requires 32k context for full training context
Best for
- Research on MoE architectures
- Benchmarking reasoning tasks
- Developing high‑quality code assistants
Who it’s for. Researchers and developers interested in efficient large‑scale language models. The model is suitable for those who need strong reasoning and coding capabilities without the full parameter budget of a dense 40B model.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 0.3 GB | 1.3 GB | Fits | Get file → | |
| IQ1_S | 21.4 GB | 22.5 GB | Too big | Get file → | |
| IQ1_M | 22.2 GB | 23.3 GB | Too big | Get file → | |
| IQ2_XXS | 25.8 GB | 26.9 GB | Too big | Get file → | |
| IQ2_XS | 28.5 GB | 29.6 GB | Too big | Get file → | |
| IQ2_S | 28.7 GB | 29.7 GB | Too big | Get file → | |
| IQ2_M | 32.4 GB | 33.4 GB | Too big | Get file → | |
| Q2_K | 36.2 GB | 37.3 GB | Too big | Get file → | |
| Q2_K_L | 36.9 GB | 37.9 GB | Too big | Get file → | |
| IQ3_XXS | 40.8 GB | 41.9 GB | Too big | Get file → | |
| IQ3_XS | 42.5 GB | 43.5 GB | Too big | Get file → | |
| Q3_K_S | 44.9 GB | 45.9 GB | Too big | Get file → | |
| IQ3_M | 47.1 GB | 48.2 GB | Too big | Get file → | |
| Q3_K_M | 47.1 GB | 48.2 GB | Too big | Get file → | |
| Q3_K_L | 49.0 GB | 50.0 GB | Too big | Get file → | |
| Q3_K_XL | 49.6 GB | 50.6 GB | Too big | Get file → | |
| IQ4_XS | 55.2 GB (2 parts) | 56.2 GB | Too big | Get file → | |
| IQ4_NL | 58.3 GB (2 parts) | 59.4 GB | Too big | Get file → | |
| Q4_0 | 59.3 GB (2 parts) | 60.3 GB | Too big | Get file → | |
| Q4_K_S | 60.4 GB (2 parts) | 61.4 GB | Too big | Get file → | |
| Q4_K_M | 62.6 GB (2 parts) | 63.7 GB | Too big | Get file → | |
| Q4_K_L | 63.1 GB (2 parts) | 64.1 GB | Too big | Get file → | |
| Q4_1 | 64.6 GB (2 parts) | 65.7 GB | Too big | Get file → | |
| Q5_K_S | 71.0 GB (2 parts) | 72.1 GB | Too big | Get file → | |
| Q5_K_M | 73.3 GB (2 parts) | 74.4 GB | Too big | Get file → | |
| Q6_K | 84.6 GB (3 parts) | 85.6 GB | Too big | Get file → | |
| Q8_0 | 109.4 GB (3 parts) | 110.5 GB | Too big | Get file → |
For 8 GB VRAM, use IQ1_S (21.45 GB file, ~22.5 GB memory at 8k context). For 16 GB VRAM, IQ2_M (32.41 GB file, ~33.4 GB memory) or Q3_K_S (44.9 GB file, ~45.9 GB memory). For 32 GB VRAM, Q4_K_M (62.62 GB file, ~63.7 GB memory) or Q5_K_S (71.03 GB file, ~72.1 GB memory).
What GPU do you need for Ling Flash 2.0?
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 the main file (0.3 GB file, about 1.3 GB in use).
- 6 GB VRAM: the best quant that fits is the main file (0.3 GB file, about 1.3 GB in use).
- 8 GB VRAM: the best quant that fits is the main file (0.3 GB file, about 1.3 GB in use).
- 12 GB VRAM: the best quant that fits is the main file (0.3 GB file, about 1.3 GB in use).
- 16 GB VRAM: the best quant that fits is the main file (0.3 GB file, about 1.3 GB in use).
- 24 GB VRAM: the best quant that fits is IQ1_M (22.2 GB file, about 23.3 GB in use).
- 32 GB VRAM: the best quant that fits is IQ2_S (28.7 GB file, about 29.7 GB in use).
- 48 GB VRAM: the best quant that fits is Q3_K_S (44.9 GB file, about 45.9 GB in use).
Which Ling Flash 2.0 quant should you choose?
Q4_K_M is the usual starting point for Ling Flash 2.0: it balances file size and quality well. Here it is a 62.6 GB file.
The Q8_0 file is 109.4 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 the first one in the table at 0.3 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 Flash 2.0
How much VRAM does Ling Flash 2.0 need?
About 63.7 GB at Q4_K_M with an 8K context window. That is the 62.6 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 Ling Flash 2.0 on an 8 GB GPU?
Yes, with the smaller files: the 0.3 GB file needs about 1.3 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Ling Flash 2.0 GGUF?
It depends on the quant: file is 0.3 GB, IQ1_S is 21.4 GB, IQ1_M is 22.2 GB, IQ2_XXS is 25.8 GB, IQ2_XS is 28.5 GB, IQ2_S is 28.7 GB and 21 more.
What context length does Ling Flash 2.0 support?
The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 0.5 GB at 8K and 2.1 GB at 32K.
Where can I download Ling Flash 2.0 GGUF?
The files are on Hugging Face in the bartowski/inclusionAI_Ling-flash-2.0-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 Flash 2.0 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.
Can I run Ling Flash 2.0 on 8 GB VRAM?
The smallest quant for 8 GB VRAM is IQ1_S, which requires 21.45 GB file size and approximately 22.5 GB memory at 8k context. It will not fit in 8 GB VRAM.
Does Ling Flash 2.0 support vision tasks?
No, the model does not include vision capabilities.