Ranked #1 on the Decision Index, with the highest Full score, 73.9, on the maintainers' private tests (edition 0.4.1, board of 12 November 2026, 300 entries).
Decision Index leaderboard, Full score, 12 November 2026.
A multimodal decision model: given a state (a web page and its screenshot, a document, a conversation: text, JSON and images) and typed questions about it, it returns a probability for every option of every question in one forward pass, without generating text. Fine-tuned from google/gemma-4-31B-it, with a linear readout over the options.
state.viewport_screenshot, or data URLs in images)choice (up to 255 options), noul (true or false), score (a level on a scale)POST /v1/systemone)The board ranks by the Full score, which its maintainers measure on private tests only: 60% tests of the same skills as the public suite and 40% tasks from new domains.
| Decision Index 0.4.1, board of 12 November 2026 | karpo-decision-31B |
|---|---|
| Full score | 73.9, rank #1, the highest of 300 entries |
| Private tests, same skills (on the public scale) | 70.7 |
| Private tasks, new domains (on the public scale) | 78.6 |
| Public index (our run, adopted by the board) | 75.0 |
| Median latency (one GPU, one request at a time) | 94 ms per request |
| Calibration error | 4.8 points |
The board gives one rank to models within 0.9 points of the first model in a group: karpo-decision-31B (73.9) and GLiDE 27B (73.4) share #1.
Our run of the official kit: 140,178 scored requests, 11 unsupported (a question over 32,768 tokens with its state). Results: runs/karpo-decision-31B.
| score | |
|---|---|
| Public index (chance-corrected) | 75.0 |
| Raw index | 80.9 |
| Breadth | 74.7 |
| Knowledge & Reasoning | 65.5 |
| Language Understanding | 76.8 |
| Retrieval & Classification | 76.4 |
| Tools & Automation | 88.4 |
| Arts & Human Taste | 68.2 |
Score vs. size on the Decision Index 0.3 public suite, board of 12 November 2026.
| Model | Base | Parameters |
|---|---|---|
| karpo-decision-31B | Gemma 4 31B | 31.3B |
| karpo-decision-12B | Gemma 4 12B | 12.0B |
| karpo-decision-9B | Qwen3.5 9B | 8.4B |
| karpo-decision-4B | Qwen3.5 4B | 4.5B |
| karpo-decision-2B | Qwen3.5 2B | 2.2B |
| karpo-decision-0.8B | Qwen3.5 0.8B | 0.85B |
Answer: ( slot that ends a
question's branch, a linear readout scores 255 option codes and a softmax over the question's own options gives the
probabilities. Nothing is decoded, so the answer is always one of the options.The question-list-first layout follows FlexDM and jev-brain; the shared-block and branch rule follows Sieve-31B-v3.
Inference code: github.com/machinepulse-ai/karpo-decision. Python 3.12 or later and one GPU; the weights take 62.5 GB, long requests more. Loading the checkpoint as a plain chat model does not reproduce its answers.
pip install "karpo-decision @ git+https://github.com/machinepulse-ai/karpo-decision"
from huggingface_hub import snapshot_download
from karpo_decision.decider import Decider
path = snapshot_download("MachinePulseAI/karpo-decision-31B")
decider = Decider.from_pretrained(path, device="cuda")
response = decider.decide({
"state": "Order 1182: the customer asks to return a blender bought 45 days ago. Policy: returns within 30 days.",
"questions": {
"eligible": {"type": "noul", "instructions": "Is the return eligible under the policy?"},
"action": {"type": "choice", "instructions": "What should the agent do?",
"criteria": {"refund": "Issue a full refund", "decline": "Politely decline the return",
"escalate": "Escalate to a supervisor"}},
},
})
print(response["answers"])
# {"eligible": {"type": "noul", "noul": P(true)},
# "action": {"type": "choice", "choice": <key>, "probabilities": {<key>: p, ...}, "confidence": c}}
HTTP server (POST /v1/systemone, GET /healthz):
CKPT=$(python -c 'from huggingface_hub import snapshot_download; print(snapshot_download("MachinePulseAI/karpo-decision-31B"))')
torchrun --standalone --nproc_per_node 1 -m karpo_decision.serve.server --checkpoint "$CKPT" --port 18440
Decision Index kit, installed separately:
pip install "decision-index @ git+https://github.com/apolinario/decision-index"
python -m decision_index run --engine karpo_decision.serve.engine:DecisionIndexEngine \
--option checkpoint="$CKPT" --out runs/karpo-decision-31B
google/gemma-4-31B → google/gemma-4-31B-it → karpo-decision-31B (fine-tuned)
The weights and model files are licensed under the Apache License, Version 2.0. Copyright 2026 MachinePulse Pte. Ltd.
karpo-decision-31B is a modified version of google/gemma-4-31B-it (Apache-2.0): decoder weights fine-tuned, readout added. Gemma 4 stays under its own licence, a copy of which is in licenses/Gemma-4-LICENSE; see NOTICE.
The inference code (github.com/machinepulse-ai/karpo-decision) is licensed under Apache-2.0.
Not required by the licence, but appreciated: if a product or service uses this model or a derivative, show
Powered by karpo-decision (MachinePulse)
with a link to https://huggingface.co/MachinePulseAI/karpo-decision-31B where the medium allows one.
model-00010-of-00012.safetensors 5.4 GBmodel-00007-of-00012.safetensors 5.4 GBmodel-00002-of-00012.safetensors 5.4 GBmodel-00008-of-00012.safetensors 5.3 GBmodel-00003-of-00012.safetensors 5.3 GBmodel-00005-of-00012.safetensors 5.3 GBmodel-00004-of-00012.safetensors 5.2 GBmodel-00009-of-00012.safetensors 5.2 GBmodel-00001-of-00012.safetensors 5.2 GBmodel-00006-of-00012.safetensors 5.2 GBmodel-00011-of-00012.safetensors 5.2 GBmodel-00012-of-00012.safetensors 4.6 GBreadout.safetensors 5 MBgoogle/gemma-4-31B-it