Artemis 31B v1.2

Q6_K + BF16 projector · bartowski · Preserved candidate · quality ungraded

bartowski's Q6_K quantization of TheDrummer Artemis 31B v1.2, preserved with the matching BF16 multimodal projector.

Benchmark results · Files · Download · Runtime · Provenance · Full collection

Join Infinity on Discord

Get model support, request models and quants, discuss benchmarks, and follow new releases and model improvements.

Join the Infinity community

All model tests, per-question scores, experiment history and scoring methods

Qwen effort study — separate full-404 cohort: medium 82.87/100 versus xhigh 84.41/100; paired difference +1.54 points (95% interval [-1.71, +4.82]). 404 collected questions, 403 valid questions graded per effort; one invalid reference annotation excluded symmetrically. The historical 202-question cross-model comparison below is unchanged. Study, timing and blind subjective results.

Preserved and documented by isilicon / Mango. The original model, modification and quantization remain the upstream creators' work. This repository adds attributed preservation and exact-configuration evaluation, with the selected source bytes unchanged.

At a glance

GB is decimal and GiB is binary. Payload size is disk/storage size, not a claim that the model fits in the same amount of VRAM. Runtime, KV cache, working memory and companions add their own requirements.

Mango evaluation

No standalone Mango quality score has been published for these exact files. This candidate is preserved, but its completed matched grade is not part of this card release.

A completed score requires a native preflight, matching frozen prompts, 202 suite answers, 12 agent cases, archive verification and the same reference graders. The benchmark methodology describes the protocol; publisher-side results are retained separately below.

Runtime and compatibility

Text, image and audio support described upstream requires the relevant runtime and projector path. No multimodal quality, context ceiling or throughput result has been measured for this selected set by Mango.

Preserved candidate; no standalone Mango score and no inference of equivalence to a Qwen or GLM result.

The local evaluation covers text/tool tasks. It does not establish the full advertised context ceiling, image/audio quality or every optional speculative-decoding path. Publisher capabilities and Mango-verified behavior are separate scopes.

Companion artifacts

No separate companion mirror is declared in this selected manifest. This does not imply that every optional upstream projector, draft head or multimodal path is included; consult the exact file list.

Files and integrity

Full paths, byte sizes and complete SHA-256/Git blob identities are in SOURCE_MANIFEST.json. Available original README/license/notice files are also preserved. MIRROR_STATUS.json records the earlier source-file verification; MODEL_CARD_STATUS.json records this documentation refresh separately.

Download and use

The commands pin the already-verified mirror weight revision. Documentation can advance on main without changing that preserved weight snapshot. Keep original subfolders and all shards.

python -m pip install -U huggingface_hub

hf download isilicon/mango-artemis-31b-q6k \
  TheDrummer_Artemis-31B-v1.2-Q6_K.gguf \
  mmproj-TheDrummer_Artemis-31B-v1.2-bf16.gguf \
  --revision c8e18b701a1f7d3dd98d311c9f7cacc05432c86d \
  --local-dir ./weights/artemis-31b-q6k

Load example and setup scope

llama-server \
  --model ./weights/artemis-31b-q6k/TheDrummer_Artemis-31B-v1.2-Q6_K.gguf \
  --host 127.0.0.1 --port 8080 --ctx-size 32768 \
  --jinja --flash-attn on --cache-type-k q8_0 --cache-type-v q8_0

This is a load-command pattern, not a validated Mango inference or performance result for this candidate. Check architecture support before loading; enable projector/MTP options only after their own preflight.

Provenance and license

The source describes a 31B multimodal model and credits TheDrummer. The Apache license provenance was checked through the pinned base chain. Mango has held this candidate out of its modified-model benchmark selection because the available cards did not establish the required refusal-modification evidence.

Quant source names TheDrummer base; pinned base names google/gemma-4-31B-it, whose pinned metadata declares Apache2. Preserve this provenance and canonical license text.

Original notices, base-model conditions and the publisher's stated limitations still apply. Downloadable preservation does not add new license rights. Source-described refusal modification and publisher-side benchmark claims have not been independently re-measured by this Mango task cohort.

Original publisher documentation

The unchanged upstream README and pinned source card retain the original usage guidance, technical discussion, attribution, caveats and citations.

Read the complete original publisher card

Source reference. The following is the original publisher's documentation. Its other quants, full-model files, context claims and benchmark tables can exceed the selection in this mirror. Relative asset/file URLs are qualified to the pinned original source; publisher measurements remain separate from Mango measurements.

Llamacpp imatrix Quantizations of Artemis-31B-v1.2 by TheDrummer

Using llama.cpp release b11159 for quantization.

Original model: https://huggingface.co/TheDrummer/Artemis-31B-v1.2

Model details:

• Parameter count: 31B
• Input support: text, image and audio (with mmproj file) - details
• Speculative decoding: no
• imatrix: yes - details

How to run

Prompt format

<bos><|turn>system
{system_prompt}<turn|>
<|turn>user
{prompt}<turn|>
<|turn>model
<|channel>thought
<channel|>

Prompt format with tool definitions

<bos><|turn>system
{system_prompt}<|tool>declaration:get_stock_price{description:<|"|>Get the current stock price<|"|>,parameters:{properties:{symbol:{description:<|"|>The stock symbol, e.g. AAPL, GOOG<|"|>,type:<|"|>STRING<|"|>}},required:[<|"|>symbol<|"|>],type:<|"|>OBJECT<|"|>}}<tool|><turn|>
<|turn>user
{prompt}<turn|>
<|turn>model
<|channel>thought
<channel|>

Don't know which to choose? Grab Q4_K_M (19.52GB) - usually a good mix of size and performance. Download instructions available here

Available files:

Download a specific file:

hf download bartowski/TheDrummer_Artemis-31B-v1.2-GGUF --include "TheDrummer_Artemis-31B-v1.2-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/TheDrummer_Artemis-31B-v1.2-GGUF --include "TheDrummer_Artemis-31B-v1.2-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/TheDrummer_Artemis-31B-v1.2-GGUF --include "TheDrummer_Artemis-31B-v1.2-bf16/*" --local-dir ./

You can either specify a new local-dir (TheDrummer_Artemis-31B-v1.2-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/TheDrummer_Artemis-31B-v1.2-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 b11159 - 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 and audio input. Alongside the quants, this repo includes the multimodal projector files mmproj-TheDrummer_Artemis-31B-v1.2-bf16.gguf and mmproj-TheDrummer_Artemis-31B-v1.2-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.

Per-tensor layouts

Some of these files were built with a layout computed for this model instead of llama.cpp's standard one-size-fits-all rules. A Q4_K_M is still mostly Q4_K; the extra precision goes to the weights this particular model is most sensitive to. The S, M or L in a name says how much of the model stays at the base precision: about 90 % for S, 70 % for M and 50 % for L. An _L name is simply the large size of its family. Q4_K_L is to Q4_K_M what Q4_K_M is to Q4_K_S; Q6_K_S, Q6_K and Q6_K_L are the small, medium and large sizes of Q6_K, with Q6_K_L about halfway to Q8_0. In earlier releases an _L name meant the embedding and output weights were kept at Q8_0; in these files it means the larger size of the base type. There is no size target, so each file's bits per weight is reported rather than promised.

The layout each of these files was built with is published in the layouts/ folder: <file>.tensor-types.txt is the exact --tensor-type-file given to llama-quantize, and <file>.layout.json records how it was computed, including the generator version, the llama.cpp release and the commit, so any of them can be rebuilt. The code that computed them is public at quantization-config; its tag key-3bf8b43e20a68f0b is the exact snapshot these files record. The method is described in this write-up.

Checked on this model before any of these files were released: Q4_K_M reached 1.01×, Q3_K_M 0.90× and IQ2_XXS 0.80× the KL divergence of the standard layout at the same file size.

Layout details

Files built from a computed layout:

Checked on this model: the computed layout against the standard one, measured by KL divergence against the unquantized model. The ratio compares each computed file with the standard ladder read at that file's own size, so it can differ from the two KLD columns when the two files differ in size.

How it works: the base type is a floor for every body tensor and a fixed share of the body bytes stays at it (90 % for S, 70 % for M, 50 % for L); the remaining bytes go where a cross-model sensitivity prior, measured by KL divergence against the unquantized model, says they buy the most quality. The embedding and output tensors are sized by their share of the file: a small table is kept at Q8_0, a large one follows the file's bitrate. A K-quant and the IQ quant with the same base bitrate (Q3_K_S and IQ3_XS, Q3_K_M and IQ3_S, Q3_K_L and IQ3_M) come out at about the same size; the IQ file is the GPU-oriented twin.

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: TheDrummer_Artemis-31B-v1.2-calibration-v6.txt. The imatrix is available here: TheDrummer_Artemis-31B-v1.2-imatrix.gguf.

Calibration render details

{
  "generator": "auto_quant_v2 calibration renderer",
  "recipe": "calibration-v6",
  "model": "Artemis-31B-v1.2",
  "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": 219,
  "tool_chunks": 273,
  "total_chunks": 492,
  "tool_chunk_fraction": 0.555,
  "n_conversations": 137,
  "extension_convs_used": 0,
  "conversation_token_lengths": [
    345,
    1411,
    652,
    1146,
    879,
    1186,
    2285,
    410,
    987,
    945,
    722,
    1728,
    667,
    739,
    2132,
    993,
    836,
    713,
    552,
    518,
    1152,
    782,
    1143,
    1012,
    1435,
    1215,
    797,
    470,
    752,
    1242,
    1309,
    590,
    1027,
    761,
    986,
    1458,
    1413,
    928,
    284,
    1746,
    1225,
    888,
    1157,
    1491,
    1715,
    1046,
    1269,
    518,
    2154,
    909,
    2418,
    569,
    761,
    515,
    919,
    491,
    2037,
    498,
    871,
    821,
    1008,
    963,
    680,
    772,
    742,
    1163,
    669,
    1272,
    1562,
    660,
    329,
    881,
    2853,
    2736,
    529,
    505,
    775,
    550,
    1061,
    869,
    921,
    526,
    807,
    826,
    926,
    1234,
    1158,
    1172,
    612,
    425,
    2180,
    660,
    907,
    1319,
    1790,
    972,
    417,
    1191,
    1133,
    1381,
    1394,
    1415,
    591,
    534,
    814,
    2647,
    530,
    357,
    561,
    982,
    872,
    1107,
    539,
    361,
    332,
    2029,
    496,
    923,
    1593,
    1438,
    2082,
    2110,
    459,
    670,
    615,
    600,
    1009,
    769,
    646,
    1093,
    547,
    517,
    1170,
    710,
    688,
    1087,
    1257
  ],
  "warnings": []
}

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.

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

Preservation record

Selected source bytes are verified against pinned metadata; the original eight graded mirrors also passed the previously recorded anonymous 1 MiB weight-prefix checks. This is not a full independent-device restore/load test. Hugging Face repository copies share a provider and do not guarantee permanent storage.

Collection · Source manifest · Original documentation