bartowski/AesCode-32B-GGUF

Llamacpp imatrix Quantizations of AesCode-32B by microsoft

Using llama.cpp release b11490 for quantization.

Original model: https://huggingface.co/microsoft/AesCode-32B

Model details:

• Parameter count: 33B (source checkpoint)
• Input support: text, image (with mmproj file) - details
• Speculative decoding: no
• imatrix: yes - details

How to run

Prompt format

<|im_start|>system
{system_prompt}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant

Prompt format with tool definitions

<|im_start|>system
{system_prompt}

# Tools

You may call one or more functions to assist with the user query.

You are provided with function signatures within <tools></tools> XML tags:
<tools>
{"type": "function", "function": {"name": "get_stock_price", "description": "Get the current stock price", "parameters": {"type": "object", "properties": {"symbol": {"type": "string", "description": "The stock symbol, e.g. AAPL, GOOG"}}, "required": ["symbol"]}}}
</tools>

For each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:
<tool_call>
{"name": <function-name>, "arguments": <args-json-object>}
</tool_call><|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant

Don't know which to choose? Grab Q4_K_M (20.49GB) - usually a good mix of size and performance. Download instructions available here

Available files:

Download a specific file:

hf download bartowski/AesCode-32B-GGUF --include "AesCode-32B-Q4_K_M.gguf" --local-dir ./

Downloading using the Hugging Face CLI

Click to view download instructions

First, make sure you have the Hugging Face CLI installed:

pip install -U "huggingface_hub[cli]"

Download a specific file:

hf download bartowski/AesCode-32B-GGUF --include "AesCode-32B-Q4_K_M.gguf" --local-dir ./

The files marked true in the Split column above are stored as multiple parts in a folder. To download all the parts to a local folder, run:

hf download bartowski/AesCode-32B-GGUF --include "AesCode-32B-bf16/*" --local-dir ./

You can either specify a new local-dir (AesCode-32B-bf16) or download them all in place (./)

How to run

These quants run with llama.cpp - installable in one line via llama.app:

curl -LsSf https://llama.app/install.sh | sh
llama-server -hf bartowski/AesCode-32B-GGUF:Q4_K_M

llama-server includes a built-in chat web UI, served at http://localhost:8080 by default.

These quants were made with llama.cpp release b11490 - if this model's architecture is newly supported, you'll need that release or newer to run them.

They also work in: LM Studio · koboldcpp · ramalama · Jan AI · Text Generation Web UI · LoLLMs · Atomic Chat

Multimodal

This model supports image input. Alongside the quants, this repo includes the multimodal projector files mmproj-AesCode-32B-bf16.gguf and mmproj-AesCode-32B-f16.gguf, which pair with any quant above.

llama.cpp downloads the mmproj automatically when using -hf as shown above; if you're loading files manually, pass it with --mmproj.

imatrix

All quants made using imatrix option, with a calibration corpus rendered through this model's own chat template. The corpus pairs plain prose with tool-calling and reasoning conversations (corpus source data), encoded exactly as this model sees them at inference and processed with --parse-special, so chat-format special tokens contribute to the importance matrix. The corpus rendered for this model is included in this repo: AesCode-32B-calibration-v6.txt. The imatrix is available here: AesCode-32B-imatrix.gguf.

Calibration render details

{
  "recipe": "calibration-v6",
  "encoder": "chat_template",
  "library_versions": {
    "transformers": "5.9.0",
    "tokenizers": "0.22.2",
    "tiktoken": "0.14.0",
    "blobfile": "3.3.0",
    "huggingface_hub": "1.31.0"
  },
  "chunk_size": 512,
  "prose_chunks": 218,
  "tool_chunks": 268,
  "total_chunks": 486,
  "n_conversations": 164,
  "conversation_token_lengths": [
    319,
    847,
    573,
    1180,
    806,
    635,
    719,
    2084,
    1087,
    410,
    612,
    967,
    810,
    1200,
    481,
    669,
    2076,
    771,
    696,
    633,
    695,
    367,
    316,
    637,
    667,
    615,
    1009,
    580,
    547,
    785,
    1530,
    807,
    637,
    477,
    622,
    1229,
    999,
    455,
    509,
    718,
    425,
    1262,
    1075,
    703,
    244,
    778,
    645,
    646,
    984,
    1545,
    879,
    1713,
    680,
    944,
    536,
    2326,
    2079,
    450,
    2395,
    489,
    666,
    456,
    332,
    377,
    1978,
    518,
    607,
    661,
    636,
    729,
    579,
    716,
    651,
    559,
    706,
    689,
    595,
    1115,
    401,
    786,
    1583,
    816,
    265,
    130,
    345,
    731,
    2760,
    1947,
    357,
    1277,
    472,
    621,
    500,
    718,
    317,
    691,
    398,
    791,
    621,
    846,
    1036,
    799,
    636,
    846,
    827,
    539,
    617,
    368,
    483,
    1158,
    2144,
    2082,
    328,
    318,
    1248,
    847,
    844,
    2071,
    1357,
    613,
    295,
    581,
    1115,
    1091,
    1406,
    1115,
    522,
    676,
    508,
    242,
    651,
    2069,
    406,
    926,
    418,
    435,
    925,
    248,
    1071,
    458,
    147,
    124,
    1977,
    452,
    650,
    1695,
    1386,
    2020,
    2057,
    467,
    636,
    497,
    528,
    620,
    563,
    358,
    786,
    548,
    404,
    1095,
    617,
    516,
    739,
    1011
  ]
}

Embed/output weights

Some of these quants (Q3_K_XL, Q4_K_L etc) are the standard quantization method with the embeddings and output weights quantized to Q8_0 instead of what they would normally default to.

ARM/AVX information

llama.cpp automatically "repacks" weights into an interleaved layout at load time for faster inference on ARM and AVX machines - details in this PR. This once required downloading special Q4_0_4_4/4_8/8_8 files; those are long gone. Online repacking now covers Q4_0, IQ4_NL, and most K-quants, so no special quant choice is needed for CPU inference.

Which file should I choose?

Click here for details

An older (early 2024) but still useful write-up with charts comparing quant performances is provided by Artefact2 here

The first thing to figure out is how big a model you can run. To do this, you'll need to figure out how much RAM and/or VRAM you have.

If you want your model running as FAST as possible, you'll want to fit the whole thing on your GPU's VRAM. Aim for a quant with a file size 1-2GB smaller than your GPU's total VRAM.

If you want the absolute maximum quality, add both your system RAM and your GPU's VRAM together, then similarly grab a quant with a file size 1-2GB Smaller than that total.

Hugging Face can also do this math for you: add your hardware in your Local Apps settings and the model page will show which files fit.

Next, you'll need to decide if you want to use an 'I-quant' or a 'K-quant'.

If you don't want to think too much, grab one of the K-quants. These are in format 'QX_K_X', like Q5_K_M.

If you want to get more into the weeds, you can check out this extremely useful feature chart:

llama.cpp feature matrix

But basically, if you're aiming for below Q4, and you're running cuBLAS (Nvidia) or rocBLAS (AMD), you should look towards the I-quants. These are in format IQX_X, like IQ3_M. These are newer and offer better performance for their size.

These I-quants can also be used on CPU, but will be slower than their K-quant equivalent, so speed vs performance is a tradeoff you'll have to decide.

Credits

Thank you kalomaze and Dampf for assistance in creating the imatrix calibration dataset.

Thank you ZeroWw for the inspiration to experiment with embed/output.

Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski