UI-Venus-2-9B-GGUF

Llamacpp imatrix Quantizations of UI-Venus-2-9B by inclusionAI

Using <a href="https://github.com/ggml-org/llama.cpp/">llama.cpp</a> release <a href="https://github.com/ggml-org/llama.cpp/releases/tag/b10630">b10630</a> for quantization.

Original model: https://huggingface.co/inclusionAI/UI-Venus-2-9B

Model details:

• Parameter count: 9B
• 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
<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/UI-Venus-2-9B-GGUF --include "UI-Venus-2-9B-Q4_K_M.gguf" --local-dir ./

Downloading using the Hugging Face CLI

<details> <summary>Click to view download instructions</summary>

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

pip install -U "huggingface_hub[cli]"

Download a specific file:

hf download bartowski/UI-Venus-2-9B-GGUF --include "UI-Venus-2-9B-Q4_K_M.gguf" --local-dir ./

</details>

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/UI-Venus-2-9B-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 b10630 - 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-UI-Venus-2-9B-f16.gguf and mmproj-UI-Venus-2-9B-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: UI-Venus-2-9B-calibration-v6.txt. The imatrix is available here: UI-Venus-2-9B-imatrix.gguf.

<details> <summary>Calibration render details</summary>

{
  "generator": "auto_quant_v2 calibration renderer",
  "recipe": "calibration-v6",
  "model": "UI-Venus-2-9B",
  "encoder": "chat_template",
  "chunk_size": 512,
  "prose_chunks": 214,
  "tool_chunks": 358,
  "total_chunks": 572,
  "tool_chunk_fraction": 0.626,
  "n_conversations": 137,
  "extension_convs_used": 0,
  "conversation_token_lengths": [
    566,
    1629,
    1259,
    1521,
    1147,
    1366,
    3182,
    828,
    1229,
    1424,
    1069,
    2098,
    883,
    1255,
    2791,
    1275,
    1122,
    988,
    739,
    720,
    1364,
    1056,
    1409,
    1204,
    1870,
    1500,
    1656,
    910,
    1417,
    1648,
    1571,
    1222,
    1270,
    1039,
    1018,
    1714,
    1653,
    1177,
    479,
    1912,
    1413,
    1128,
    1400,
    2000,
    2092,
    1313,
    1611,
    909,
    2958,
    1104,
    2866,
    789,
    1026,
    972,
    905,
    700,
    2492,
    917,
    1150,
    1089,
    1229,
    1173,
    924,
    1221,
    1185,
    1590,
    915,
    1554,
    2124,
    843,
    323,
    1124,
    3337,
    2838,
    711,
    944,
    1030,
    1075,
    1296,
    1084,
    1149,
    797,
    1212,
    1073,
    1293,
    1548,
    1405,
    2068,
    875,
    660,
    2768,
    640,
    1405,
    1677,
    1949,
    1205,
    645,
    1358,
    1122,
    1731,
    1846,
    1717,
    819,
    1017,
    1039,
    2858,
    742,
    732,
    765,
    1414,
    1054,
    1579,
    770,
    350,
    319,
    2628,
    1020,
    1141,
    1828,
    2041,
    2697,
    2679,
    818,
    982,
    842,
    961,
    1237,
    1001,
    844,
    1359,
    832,
    712,
    1745,
    1029,
    919,
    1318,
    1481
  ],
  "warnings": []
}

</details>

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?

<details> <summary>Click here for details</summary>

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.

</details>

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