Jackdaw 3 Ministral needs about 13.2 GB of memory at Q4_K_M (a 11.8 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 24 quants, from 0.0 GB to 15.9 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 3.5 GB.
Jackdaw 3 Ministral is a 19.3B-parameter Mistral model in GGUF format, quantized by mradermacher. Its trained context length is 256K tokens. The repository has 24 quant files, from 0.0 GB to 15.9 GB (Q6_K).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 0.0 GB | 1.4 GB | Fits | Get file → | |
| IQ1_S | 4.3 GB | 5.7 GB | Fits | Get file → | |
| IQ1_M | 4.7 GB | 6.1 GB | Fits | Get file → | |
| IQ2_XXS | 5.4 GB | 6.8 GB | Fits | Get file → | |
| IQ2_XS | 6.0 GB | 7.3 GB | Tight | Get file → | |
| IQ2_S | 6.1 GB | 7.4 GB | Tight | Get file → | |
| IQ2_M | 6.6 GB | 8.0 GB | Tight | Get file → | |
| Q2_K_S | 6.8 GB | 8.2 GB | Too big | Get file → | |
| Q2_K | 7.4 GB | 8.7 GB | Too big | Get file → | |
| IQ3_XXS | 7.7 GB | 9.1 GB | Too big | Get file → | |
| IQ3_XS | 8.2 GB | 9.6 GB | Too big | Get file → | |
| IQ3_S | 8.6 GB | 10.0 GB | Too big | Get file → | |
| Q3_K_S | 8.6 GB | 10.0 GB | Too big | Get file → | |
| IQ3_M | 9.0 GB | 10.4 GB | Too big | Get file → | |
| Q3_K_M | 9.5 GB | 10.8 GB | Too big | Get file → | |
| Q3_K_L | 10.1 GB | 11.5 GB | Too big | Get file → | |
| IQ4_XS | 10.5 GB | 11.9 GB | Too big | Get file → | |
| Q4_0 | 11.1 GB | 12.5 GB | Too big | Get file → | |
| Q4_K_S | 11.2 GB | 12.5 GB | Too big | Get file → | |
| Q4_K_M | 11.8 GB | 13.2 GB | Too big | Get file → | |
| Q4_1 | 12.2 GB | 13.6 GB | Too big | Get file → | |
| Q5_K_S | 13.4 GB | 14.8 GB | Too big | Get file → | |
| Q5_K_M | 13.8 GB | 15.2 GB | Too big | Get file → | |
| Q6_K | 15.9 GB | 17.2 GB | Too big | Get file → |
What GPU do you need for Jackdaw 3 Ministral?
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 (0.0 GB file, about 1.4 GB in use).
- 6 GB VRAM: the best quant that fits is IQ1_S (4.3 GB file, about 5.7 GB in use).
- 8 GB VRAM: the best quant that fits is IQ2_M (6.6 GB file, about 8.0 GB in use).
- 12 GB VRAM: the best quant that fits is IQ4_XS (10.5 GB file, about 11.9 GB in use).
- 16 GB VRAM: the best quant that fits is Q5_K_M (13.8 GB file, about 15.2 GB in use).
- 24 GB VRAM: the best quant that fits is Q6_K (15.9 GB file, about 17.2 GB in use).
Which Jackdaw 3 Ministral quant should you choose?
Q4_K_M is the usual starting point for Jackdaw 3 Ministral: it balances file size and quality well. Here it is a 11.8 GB file.
If memory is tight, the smallest file is the first one in the table at 0.0 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 Jackdaw 3 Ministral
How much VRAM does Jackdaw 3 Ministral need?
About 13.2 GB at Q4_K_M with an 8K context window. That is the 11.8 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 Jackdaw 3 Ministral on an 8 GB GPU?
Yes, with IQ2_M: the 6.6 GB file needs about 8.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Jackdaw 3 Ministral GGUF?
It depends on the quant: file is 0.0 GB, IQ1_S is 4.3 GB, IQ1_M is 4.7 GB, IQ2_XXS is 5.4 GB, IQ2_XS is 6.0 GB, IQ2_S is 6.1 GB and 18 more.
What context length does Jackdaw 3 Ministral support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.9 GB at 8K and 3.5 GB at 32K.
Where can I download Jackdaw 3 Ministral GGUF?
The files are on Hugging Face in the mradermacher/Jackdaw-3-Ministral-i1-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.