inclusionAI.Ring Flash 2.0 needs about 63.5 GB of memory at Q4_K_M (a 62.4 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 9 quants, from 23.3 GB (IQ1_M) to 109.4 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 2.1 GB.
Ring-flash-2.0: a 103B parameter MoE LLM with 32K context
Ring-flash-2.0 is a 103B parameter mixture of experts model based on the bailingmoe2 architecture. It supports a 32768 token context window and is released under the MIT license. The model is optimized for complex reasoning, math, and code generation tasks.
Good at
- Complex reasoning
- Math competitions
- Code generation
Limits
- No vision support
- Requires large VRAM
- MIT license limits commercial use
Best for
- Research on MoE models
- Benchmarking reasoning tasks
- Developing reasoning‑heavy applications
Who it’s for. Researchers and developers who need a high‑performance MoE LLM for reasoning tasks. The model is suitable for environments that can accommodate its memory requirements and comply with the MIT license.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ1_M | 23.3 GB | 24.3 GB | Too big | Get file → | |
| IQ2_XXS | 27.1 GB | 28.2 GB | Too big | Get file → | |
| IQ2_S | 30.6 GB | 31.6 GB | Too big | Get file → | |
| Q2_K | 37.7 GB | 38.7 GB | Too big | Get file → | |
| Q3_K_M | 49.3 GB (4 parts) | 50.4 GB | Too big | Get file → | |
| Q4_K_M | 62.4 GB (5 parts) | 63.5 GB | Too big | Get file → | |
| Q5_K_M | 73.2 GB (6 parts) | 74.2 GB | Too big | Get file → | |
| Q6_K | 84.5 GB (7 parts) | 85.5 GB | Too big | Get file → | |
| Q8_0 | 109.4 GB (8 parts) | 110.5 GB | Too big | Get file → |
For 16 GB VRAM, use IQ1_M (24.3 GB memory at 8K context). For 32 GB VRAM, IQ2_XXS (28.2 GB) is appropriate. For 64 GB VRAM, IQ4_K_M (63.5 GB) or higher quants can be used. Estimates are approximate and depend on context size.
What GPU do you need for inclusionAI.Ring 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: 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: no quant fits at an 8K context.
- 32 GB VRAM: the best quant that fits is IQ2_S (30.6 GB file, about 31.6 GB in use).
- 48 GB VRAM: the best quant that fits is Q2_K (37.7 GB file, about 38.7 GB in use).
Which inclusionAI.Ring Flash 2.0 quant should you choose?
Q4_K_M is the usual starting point for inclusionAI.Ring Flash 2.0: it balances file size and quality well. Here it is a 62.4 GB file.
The Q8_0 file is 109.4 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 IQ1_M at 23.3 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 inclusionAI.Ring Flash 2.0
How much VRAM does inclusionAI.Ring Flash 2.0 need?
About 63.5 GB at Q4_K_M with an 8K context window. That is the 62.4 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 inclusionAI.Ring Flash 2.0 on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 24.3 GB, so you would need a larger GPU or CPU offloading.
What is the file size of inclusionAI.Ring Flash 2.0 GGUF?
It depends on the quant: IQ1_M is 23.3 GB, IQ2_XXS is 27.1 GB, IQ2_S is 30.6 GB, Q2_K is 37.7 GB, Q3_K_M is 49.3 GB, Q4_K_M is 62.4 GB and 3 more.
What context length does inclusionAI.Ring 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 inclusionAI.Ring Flash 2.0 GGUF?
The files are on Hugging Face in the DevQuasar-11/inclusionAI.Ring-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 inclusionAI.Ring 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 Ring-flash-2.0 on 8 GB VRAM?
No, the smallest quant requires at least 24.3 GB of VRAM at 8K context.
Does the model support vision tasks?
No, the model has no vision capabilities.
Is the model free to use for commercial projects?
It is released under the MIT license, which allows commercial use but requires attribution.