tencent_UI-Mate-9B-GGUF

Llamacpp imatrix Quantizations of UI-Mate-9B by tencent#

Using llama.cpp release b10472 for quantization.

Original model: https://huggingface.co/tencent/UI-Mate-9B

Model details:

• Parameter count: 9B
• Input support: text, image (with mmproj file) - details
• Speculative decoding: no
• imatrix: yes - details
• Perplexity/KLD measured: no

How to run

Prompt format#

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

Don't know which to choose? Grab Q4_K_M (5.91GB) - usually a good mix of size and performance. Download instructions available here

Available files:#

Download a specific file:

hf download bartowski/tencent_UI-Mate-9B-GGUF --include "tencent_UI-Mate-9B-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/tencent_UI-Mate-9B-GGUF --include "tencent_UI-Mate-9B-Q4_K_M.gguf" --local-dir ./

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/tencent_UI-Mate-9B-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 b10472 - 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-tencent_UI-Mate-9B-f16.gguf and mmproj-tencent_UI-Mate-9B-bf16.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: tencent_UI-Mate-9B-calibration-v6.txt. The imatrix is available here: tencent_UI-Mate-9B-imatrix.gguf.

Calibration render details

{
  "generator": "auto_quant_v2 calibration renderer",
  "recipe": "calibration-v6",
  "model": "UI-Mate-9B",
  "encoder": "chat_template",
  "chunk_size": 512,
  "prose_chunks": 214,
  "tool_chunks": 336,
  "total_chunks": 550,
  "tool_chunk_fraction": 0.611,
  "n_conversations": 137,
  "extension_convs_used": 0,
  "conversation_token_lengths": [
    566,
    1629,
    916,
    1404,
    1147,
    1366,
    2568,
    666,
    1229,
    1234,
    1069,
    2098,
    883,
    978,
    2402,
    1275,
    1122,
    988,
    739,
    720,
    1364,
    1060,
    1409,
    1204,
    1720,
    1500,
    1052,
    724,
    974,
    1505,
    1571,
    842,
    1270,
    1039,
    1018,
    1714,
    1653,
    1177,
    479,
    1912,
    1413,
    1128,
    1400,
    1772,
    1963,
    1317,
    1611,
    783,
    2443,
    1104,
    2639,
    789,
    1026,
    753,
    905,
    700,
    2315,
    760,
    1150,
    1089,
    1233,
    1173,
    928,
    1024,
    970,
    1425,
    915,
    1554,
    1837,
    843,
    323,
    1124,
    3003,
    2838,
    711,
    764,
    1034,
    797,
    1300,
    1084,
    1153,
    797,
    1073,
    1073,
    1293,
    1548,
    1405,
    1391,
    875,
    660,
    2472,
    640,
    1163,
    1677,
    1949,
    1205,
    645,
    1358,
    1126,
    1731,
    1700,
    1717,
    819,
    772,
    1039,
    2858,
    742,
    598,
    765,
    1263,
    1054,
    1372,
    770,
    350,
    319,
    2323,
    735,
    1141,
    1832,
    1730,
    2365,
    2375,
    693,
    982,
    842,
    852,
    1241,
    1001,
    844,
    1359,
    832,
    712,
    1423,
    1029,
    919,
    1318,
    1481
  ],
  "warnings": []
}

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