This is a decision model: use llama-server's /v1/systemone endpoint with -ub 4096 -b 4096. Follow Run it for supported settings and app compatibility.
In Hugging Face's generated "Use this model" commands, :BF16 and :Q8_0 name the vision projector files, not model weights. Use :Q4_K_S or the exact filename below.
Start here. Download pplx-decider-27B-Artemis-Q4_K_S.gguf (15.28 GB). 96.8% same answers as BF16 on A100; the smaller variant is very tight on a 16 GB GPU. Details.
Typed-decision accuracy on the card's selection panel, with the measured nearby-size alternatives and BF16 original. These comparisons informed file selection and do not establish performance on unseen inputs. Hardware, runtime and other evaluation panels are listed under How it was tested. The plain Q4_K_M comparator was made locally with stock llama-quantize; it is not a public third-party release.
These comparisons informed file selection; they do not establish performance on unseen inputs.
Earlier file names still work for now; new downloads should use the names below.
llama-server -hf turintech/pplx-decider-27B-Artemis-GGUF:Q4_K_S --port 8080 -c 16384 -ub 4096 -b 4096 --no-mmproj
pplx-decider is Perplexity's 27B decision model. The model accepts text (and optionally images) with a set of typed questions, and it returns a probability for every allowed answer in a single pass, instead of generating text. See the original model card.
This is an independent TurinTech conversion, not affiliated with or endorsed by Perplexity. The files are standard
GGUF and run on stock llama.cpp. For images, use the vision projector in this repository (see below). Changes from
the original: the weights are converted to GGUF and quantized for llama.cpp. See NOTICE.
| Variant | Size | Saved | Same answer | Acc. Δ | Conf. changes | Choose for |
|---|---|---|---|---|---|---|
| BF16 original | 53.8 GB | reference | reference | |||
| Q4_K_S | 15.28 GB | 72% | 96.8% | +0.3 | 0 | Default. |
| IQ3_M | 14.27 GB | 73% | 96.2% | +0.7 | 0 | Slightly less memory (15.9 GiB). |
"Saved" is the file-size reduction versus the BF16 original. "Acc. Δ" is the accuracy change in percentage points; the other columns are defined below. Variant links download the exact GGUF; full filenames are in the commands and checksums. File labels are size classes, explained under "File names, sizes and SHA-256 checksums" at the end of the card.
"Same answer as the original" is how often a file picks the same option as the BF16 original on 2,000 test decisions. "Accuracy change" is the difference in accuracy against the original on the same cases, in points. "Confident answers changed" counts answers the original was more than 80% sure of that the file changed.
As of 13 October 2026, no other public GGUF of this model runs on stock llama.cpp: Livesport's Q5_K_M and Q8_0 use their own runtime and do
not load in stock llama-server. So the comparison is with plain Q8_0 and Q4_K_M files we made from the original with
stock llama-quantize: the Q8_0 (28.6 GB) gives the same answer 99.2% of the time, the Q4_K_M (16.55 GB) 95.7%. Test and held-out results and file sizes are in the tables. The stock Q4_K_M is
closer on the benchmark sample (98.0% against 96.0%).
echo "My Stripe integration keeps failing. Please help ASAP." | ollama run turintech/pplx-decider-27b-artemis:15.3gb --ask "Does this message express urgency?"
This verified tag selects pplx-decider-27B-Artemis-Q4_K_S.gguf. It is a decision model: use /v1/systemone, not a chat prompt. Ollama 0.40.4 was tested; include "model": "turintech/pplx-decider-27b-artemis:15.3gb" in the decision request to port 11434. The command runs the documented decision example; the explicit size tag selects the recommended file and is not an immutable revision.
Needs llama.cpp from 9 October 2026 or newer. Tested at commit 1e6f04a, and both files also on release b11572
(12 October 2026), which gave identical answers on a 100-decision check.
llama-server -m pplx-decider-27B-Artemis-Q4_K_S.gguf --port 8080 -c 16384 -ub 4096 -b 4096
Tested on an Apple M3 Max (36 GB) with llama.cpp Metal (b11617): every file here loads and answers with the command above.
Keep -ub 4096 -b 4096: the numbers on this card were measured with them, and the batch size moves the
probabilities slightly. In our test on an NVIDIA RTX 5090, pplx-decider-27B-Artemis-Q4_K_S.gguf gave 0.876 at the default of 512
and 0.897 at 4096 on a 923-token input. Inputs longer than the batch size are accepted: with llama.cpp
1e6f04a and b11609, a 4,983-token input was answered at both settings. -c 16384 stops the
server from reserving memory for the full training context.
For concurrent clients, an optional configuration is -np 8 -c 32768 --cache-ram 0: eight request slots with 4,096 tokens each and the host prompt cache disabled. In our NVIDIA RTX 5090 test with llama.cpp fcac190, this used about 1.5 GiB more GPU memory than the command above. Measure the configuration on the intended workload before choosing it.
llama-server -m pplx-decider-27B-Artemis-Q4_K_S.gguf --port 8080 -np 8 -c 32768 -ub 4096 -b 4096 --cache-ram 0
curl http://localhost:8080/v1/systemone -H 'Content-Type: application/json' -d '{
"state": "My Stripe integration keeps failing. Please help ASAP.",
"questions": {"urgency": {"type": "noul", "instructions": "Does this message express urgency?"}}
}'
Questions, problems or a model to consider for a smaller version? Reply in our feedback thread.
Perplexity does not publish a llama.cpp vision projector, so this repository includes one converted from the original
weights (mmproj-pplx-decider-27B-BF16.gguf). Load it with --mmproj and raise the batch size, since several large
images can exceed 4,096 tokens:
hf download turintech/pplx-decider-27B-Artemis-GGUF mmproj-pplx-decider-27B-BF16.gguf --local-dir .
llama-server -m pplx-decider-27B-Artemis-Q4_K_S.gguf --port 8080 --mmproj mmproj-pplx-decider-27B-BF16.gguf -c 16384 -ub 8192 -b 8192
Send images as data URLs in a top-level images array of the /v1/systemone request. A smaller Q8_0 projector
(mmproj-pplx-decider-27B-Q8_0.gguf, 0.63 GB) is also included; with the original model it gives the same answer as
the BF16 projector on 98.8% of our image questions. Our image results use the BF16 projector.
This is a decision model. Run it on llama-server's /v1/systemone endpoint and keep -ub 4096 -b 4096, as in
Run it. The llama.cpp, Ollama and LM Studio buttons under "Use this model" start it as a chat model
without those settings, which is not a supported way to run it.
Download pplx-decider-27B-Artemis-Q4_K_S.gguf (15.28 GB, peak GPU memory 16.6 GiB on an NVIDIA A100,
so it should fit a 24 GB GPU) and start it with the settings below, then send requests as shown in Run it.
It gives the same answer as the original on 96.8% of 2,000 test decisions and changed no answer the original was more
than 80% sure of on the A100 (1 of 2,000 on an RTX 5090). The Artemis IQ3_M file needs 15.9 GiB, which is very tight on a
16 GB GPU (not tested on one).
These files are per-tensor mixes, and their labels are size classes (see the file names table at the end). Use the exact filename or the :Q4_K_S tag; without a tag, app buttons may select a different file.
perplexity-ai/pplx-decider-v1.1-27b is 53.8 GB and needs 50.8 GiB of GPU memory, so an 80 GB data-centre card. The default file here is 15.28 GB, 72% smaller, and needs 16.6 GiB, so it should fit a 24 GB GPU; the Artemis IQ3_M file needs 15.9 GiB (both measured on an NVIDIA A100). The default gives the same answer as the original on 96.8% of 2,000 test decisions, has measured accuracy (70.8% against 70.5%), and changed no answer the original was more than 80% sure of, in text or images (on the A100; on an RTX 5090, 1 of 2,000).
The separate train-split results are labelled held-out in the tables. Their independence from calibration and variant selection has not been verified for every file.
Test results informed candidate selection, so they do not independently measure performance on new inputs. Read them alongside the separate train-split results and public-file comparisons. Agreement measures whether a compressed model keeps the original's answers, including its mistakes. The typed-decisions accuracy figures compare answers with teacher-generated labels.
Every file was compared with the BF16 original and with plain Q8_0 and Q4_K_M quantizations, all at identical llama.cpp settings, on:
Measured with llama.cpp 1e6f04a at the settings above, every file on one NVIDIA A100 80 GB, images with the BF16
projector.
| Check | BF16 | Artemis Q4_K_S | Artemis IQ3_M | Plain Q8_0 | Plain Q4_K_M |
|---|---|---|---|---|---|
| Typed-decision accuracy | 70.5 | 70.8 | 71.2 | 70.4 | 70.7 |
| Same answer as BF16 | 96.8 | 96.2 | 99.2 | 95.7 | |
| KL to BF16 (lower is closer) | .0048 | .0080 | .0003 | .0069 | |
| Held-out accuracy | 71.0 | 70.9 | 71.2 | 70.9 | 71.0 |
| Held-out: same answer as BF16 | 97.3 | 96.0 | 99.5 | 94.9 | |
| Decision Index sample accuracy | 83.7 | 82.3 | 81.7 | 83.7 | 83.3 |
| Decision Index sample: same answer as BF16 | 96.0 | 95.3 | 100.0 | 98.0 | |
| Image accuracy | 72.1 | 72.1 | 72.1 | 72.1 | 72.4 |
| Images: same answer as BF16 | 97.6 | 96.4 | 98.8 | 96.7 | |
| Changed answers where BF16 was over 80% sure | 0 | 0 | 0 | 0 | |
| Answers changed by shuffling options (%) | 8.4 | 9.6 | 7.9 | 7.8 | 9.3 |
The original itself changes 8.4% of its answers when the options are shuffled; the compressed files stay close to that. Repeat runs gave identical answers for every file. Small differences in this sample should not be treated as general improvements.
The same files on one NVIDIA RTX 5090 32 GB (llama.cpp 1e6f04a, CUDA, -c 16384 -ub 4096 -b 4096 --parallel 2), text
checks only. The original does not fit 32 GB and ran with 35 of 65 layers on the GPU.
| File | Same answer as BF16 | Accuracy change | Confident answers changed | Held-out: same answer | Peak GPU memory |
|---|---|---|---|---|---|
| pplx-decider-27B-Artemis-Q4_K_S | 96.5% | +0.5 | 1 | 97.1% | 16.7 GiB |
| pplx-decider-27B-Artemis-IQ3_M | 95.9% | +0.5 | 0 | 96.1% | 15.9 GiB |
| Plain Q4_K_M | 96.0% | +0.1 | 0 | 95.3% | 17.8 GiB |
Within 1.5 points of the A100 on every check. On this GPU the Artemis Q4_K_S file changed 1 answer the original was more than 80% sure of (at 84%), and our two files gave identical answers on 99.6% and 99.2% of a repeat run (100% on the A100).
Peak GPU memory of llama-server on an NVIDIA A100 80 GB: text (-ub 4096 -b 4096) and with the BF16 vision projector
(-ub 8192 -b 8192).
| File | Size | Peak GPU memory, text | With images |
|---|---|---|---|
| Original, BF16 | 53.8 GB | 50.8 GiB | 53.1 GiB |
| pplx-decider-27B-Artemis-Q4_K_S | 15.28 GB | 16.6 GiB | 18.9 GiB |
| pplx-decider-27B-Artemis-IQ3_M | 14.27 GB | 15.9 GiB | 18.2 GiB |
| Plain Q4_K_M | 16.55 GB | 17.8 GiB | 20.1 GiB |
File names end in a standard llama.cpp quant label, so llama.cpp, Ollama and LM Studio can pick a file by tag. Each file is a per-tensor mix of quant types, so the label is a size class: the smallest standard llama.cpp quant whose typical bits per weight are at least the file's. The general.file_type field inside the files reads Q4_K_M, the base type given to llama-quantize; the table shows the tensor types each file holds.
| File | Earlier name | Bits per weight | Main tensor types (share of bytes) |
|---|---|---|---|
pplx-decider-27B-Artemis-Q4_K_S.gguf | pplx-decider-27B-Artemis-15.3GB.gguf | 4.54 | IQ4_XS 60%, Q4_K 27%, Q6_K 14% |
pplx-decider-27B-Artemis-IQ3_M.gguf | pplx-decider-27B-Artemis-14.3GB.gguf | 4.24 | IQ4_XS 60%, Q4_K 23%, Q6_K 8%, IQ3_XXS 3% |
The earlier names still download the same bytes, with the same checksums, for a transition period.
| File | Bytes | sha256 |
|---|---|---|
pplx-decider-27B-Artemis-Q4_K_S.gguf | 15,277,314,144 | 9e23efec609fa777dc216d9b51f3e74bb473f6aa4163c2be8bd87f0ef5e7808a |
pplx-decider-27B-Artemis-IQ3_M.gguf | 14,273,855,584 | cbc9fb46015e65fb421f8076fe9310c3413edb62d2d2802a00765f516089cb2a |
mmproj-pplx-decider-27B-BF16.gguf | 931,146,208 | 457ff3cc68b4b1cb8ba0e8383664454a838736b1ad1a4726504bcff1fb45ac2d |
mmproj-pplx-decider-27B-Q8_0.gguf | 629,247,328 | 633ea0419e6cdfca52ca0d243cbbee393eb0b03afc84f720ad5ff7c2c0e03f6c |
We also make fine-tuned and further optimised versions on request: tuned on specific decisions and data, built for the target hardware (CPU servers, laptops, Apple Silicon, edge devices), or compressed from a private model. Book a call with TurinTech.
Apache-2.0 (see LICENSE and NOTICE), the same as
perplexity-ai/pplx-decider-v1.1-27b, which is
post-trained from Qwen/Qwen3.8-27B. The original's NOTICE is reproduced in
ours.
When serving these files to others, for example through an API, an inference provider or OpenRouter, we'd appreciate a mention of TurinTech (turintech.ai) as the source of the quantization.
These files keep improving.