RakutenAI 3.0 needs about 406.1 GB of memory at Q4_K_M (a 404.5 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 10 quants, from 244.1 GB (Q2_K) to 550.9 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.
RakutenAI 3.0 is a 671B-parameter Rakutenai model in GGUF format, quantized by mmnga-o. Its trained context length is 160K tokens. The repository has 10 quant files, from 244.1 GB (Q2_K) to 550.9 GB (Q6_K).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 244.1 GB (24 parts) | 245.7 GB | Too big | Get file → | |
| Q3_K_S | 289.2 GB (30 parts) | 290.7 GB | Too big | Get file → | |
| Q3_K_M | 319.3 GB (30 parts) | 320.9 GB | Too big | Get file → | |
| Q3_K_L | 347.5 GB (35 parts) | 349.1 GB | Too big | Get file → | |
| MXFP4_MOE | 365.4 GB (36 parts) | 366.9 GB | Too big | Get file → | |
| Q4_K_S | 380.1 GB (36 parts) | 381.7 GB | Too big | Get file → | |
| Q4_K_M | 404.5 GB (40 parts) | 406.1 GB | Too big | Get file → | |
| Q5_K_S | 461.8 GB (44 parts) | 463.4 GB | Too big | Get file → | |
| Q5_K_M | 475.4 GB (51 parts) | 477.0 GB | Too big | Get file → | |
| Q6_K | 550.9 GB (59 parts) | 552.5 GB | Too big | Get file → |
What GPU do you need for RakutenAI 3.0?
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: no quant fits at an 8K context.
- 48 GB VRAM: no quant fits at an 8K context.
Which RakutenAI 3.0 quant should you choose?
Q4_K_M is the usual starting point for RakutenAI 3.0: it balances file size and quality well. Here it is a 404.5 GB file.
If memory is tight, the smallest file is Q2_K at 244.1 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 RakutenAI 3.0
How much VRAM does RakutenAI 3.0 need?
About 406.1 GB at Q4_K_M with an 8K context window. That is the 404.5 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 RakutenAI 3.0 on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 245.7 GB, so you would need a larger GPU or CPU offloading.
What is the file size of RakutenAI 3.0 GGUF?
It depends on the quant: Q2_K is 244.1 GB, Q3_K_S is 289.2 GB, Q3_K_M is 319.3 GB, Q3_K_L is 347.5 GB, MXFP4_MOE is 365.4 GB, Q4_K_S is 380.1 GB and 4 more.
What context length does RakutenAI 3.0 support?
The GGUF header lists a trained context length of 163,840 tokens (160K). The KV cache grows with the context you use: about 1.1 GB at 8K and 4.3 GB at 32K.
Where can I download RakutenAI 3.0 GGUF?
The files are on Hugging Face in the mmnga-o/RakutenAI-3.0-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 RakutenAI 3.0 use?
Hugging Face lists the license as unknown. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.