aya-23-35B needs about 32.8 GB of memory at Q4_K_M (a 21.5 GB file) with an 8K context window, so it fits on a 48 GB GPU. This repository has 23 quants, from 8.5 GB (IQ1_S) to 70.0 GB (F16).
aya-23-35B: 35B parameter command-r model in GGUF format
aya-23-35B is a 35B parameter command-r model released under cc‑by‑nc‑4.0. It is available in multiple quantizations from IQ1_S to Q8_0 and F16. The model supports 8192 context tokens and is optimized for llama.cpp.
Good at
- Large context handling
- Command‑style generation
- High‑parameter count
Limits
- No vision support
- cc‑by‑nc‑4.0 license
- Requires GPU or CPU with sufficient RAM
Best for
- Long‑form content creation
- Complex reasoning tasks
- Fine‑tuning on custom data
Who it’s for. Developers and researchers who need a large command‑style model for text generation and reasoning. The model is suitable for those who can provide the required GPU or system RAM and comply with the cc‑by‑nc‑4.0 license.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ1_S | 8.5 GB | 19.8 GB | Too big | Get file → | |
| IQ1_M | 9.1 GB | 20.4 GB | Too big | Get file → | |
| IQ2_XXS | 10.2 GB | 21.4 GB | Too big | Get file → | |
| IQ2_XS | 11.1 GB | 22.3 GB | Too big | Get file → | |
| IQ2_S | 11.8 GB | 23.1 GB | Too big | Get file → | |
| IQ2_M | 12.7 GB | 23.9 GB | Too big | Get file → | |
| Q2_K | 13.8 GB | 25.1 GB | Too big | Get file → | |
| IQ3_XXS | 13.8 GB | 25.1 GB | Too big | Get file → | |
| IQ3_XS | 15.1 GB | 26.3 GB | Too big | Get file → | |
| IQ3_S | 15.9 GB | 27.1 GB | Too big | Get file → | |
| Q3_K_S | 15.9 GB | 27.1 GB | Too big | Get file → | |
| IQ3_M | 16.7 GB | 27.9 GB | Too big | Get file → | |
| Q3_K_M | 17.6 GB | 28.9 GB | Too big | Get file → | |
| Q3_K_L | 19.1 GB | 30.4 GB | Too big | Get file → | |
| IQ4_XS | 19.2 GB | 30.4 GB | Too big | Get file → | |
| IQ4_NL | 20.2 GB | 31.5 GB | Too big | Get file → | |
| Q4_K_S | 20.4 GB | 31.6 GB | Too big | Get file → | |
| Q4_K_M | 21.5 GB | 32.8 GB | Too big | Get file → | |
| Q5_K_S | 24.3 GB | 35.6 GB | Too big | Get file → | |
| Q5_K_M | 25.0 GB | 36.2 GB | Too big | Get file → | |
| Q6_K | 28.7 GB | 39.9 GB | Too big | Get file → | |
| Q8_0 | 37.2 GB | 48.4 GB | Too big | Get file → | |
| F16 | 70.0 GB (2 parts) | 81.2 GB | Too big | Get file → |
For 8 GB VRAM, use IQ1_S (8.52 GB, 19.8 GB memory at 8K context). For 16 GB VRAM, IQ2_S (11.84 GB, 23.1 GB memory) is a good balance. For 24 GB VRAM, IQ3_S (15.86 GB, 27.1 GB memory) offers higher quality. Use Q5_K_M (25.01 GB, 36.2 GB memory) if you have 48 GB total system RAM and GPU VRAM combined.
What GPU do you need for aya-23-35B?
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 IQ2_M (12.7 GB file, about 23.9 GB in use).
- 32 GB VRAM: the best quant that fits is Q4_K_S (20.4 GB file, about 31.6 GB in use).
- 48 GB VRAM: the best quant that fits is Q6_K (28.7 GB file, about 39.9 GB in use).
Which aya-23-35B quant should you choose?
Q4_K_M is the usual starting point for aya-23-35B: it balances file size and quality well. Here it is a 21.5 GB file.
The Q8_0 file is 37.2 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 IQ1_S at 8.5 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 aya-23-35B
How much VRAM does aya-23-35B need?
About 32.8 GB at Q4_K_M with an 8K context window. That is the 21.5 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 aya-23-35B on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 19.8 GB, so you would need a larger GPU or CPU offloading.
What is the file size of aya-23-35B GGUF?
It depends on the quant: IQ1_S is 8.5 GB, IQ1_M is 9.1 GB, IQ2_XXS is 10.2 GB, IQ2_XS is 11.1 GB, IQ2_S is 11.8 GB, IQ2_M is 12.7 GB and 17 more.
What context length does aya-23-35B support?
The GGUF header lists a trained context length of 8,192 tokens (8K). The KV cache grows with the context you use: about 10.7 GB at 8K and 10.7 GB at 8K.
Where can I download aya-23-35B GGUF?
The files are on Hugging Face in the bartowski/aya-23-35B-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 aya-23-35B use?
Hugging Face lists the license as cc-by-nc-4.0. 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 aya-23-35B on 8 GB VRAM?
Yes, the IQ1_S quant is 8.52 GB and requires about 19.8 GB memory at 8K context, which fits on 8 GB VRAM with CPU fallback.
What quant should I use for 16 GB VRAM?
IQ2_S (11.84 GB, 23.1 GB memory) is recommended for 16 GB VRAM, balancing size and performance.
What license applies to aya-23-35B?
The model is licensed under cc‑by‑nc‑4.0, allowing non‑commercial use with attribution.