Decisio
v0.11.1 Release notes

Typed decisions, on your own hardware

Jev-level decisions on your own hardware, open source

Typed questions go in. A probability for every option comes out, in one pass.

  • 81 ms for one question about a new 300-token text
  • 1.08 s for 50 questions about a 3,000-token document
  • 32,000 tokens in one document
ollama pull aminroudaki/decisio-gemma
Get startedSee the proofGitHub
recordedGemma 4 31B, decisio 0.9.0, RTX PRO 6000 at 600 W
Support ticketModerationForm checkTool decision

Incoming support ticket · record 1 of 60

I was charged twice for October. Both charges are 49.00 EUR, same day. Please refund one.

reads it once91.2 ms

urgentyes/no

Does this ticket need a response within the hour?

  • yes0.16

queuechoicebilling

Which team should handle this ticket?

  • billing0.96
  • bug0.02
  • access0.01
  • +3 more

impactscorelevel 0.9

Rate the business impact using only the reported facts.

  • 00.26
  • 10.65
  • 20.05
  • 30.04

3 of 3 answered · 91.2 ms

On the board

What the public board says

  • 2ndof 128 open-weights systems on JevBench v1.6.1
  • 70.4on the intelligence axis, against Jev's 63.6, on JevBench v1.6.1
  • 73.3on the sealed questions, never published, against Jev's 61.6, on JevBench v1.6.1

Level: the two 95% intervals overlap. JevBench v1.6.1, decisio v0.8.0 rows.

See the board

Second among open-weights systems on JevBench, level with Jev, and the highest-ranked with frozen weights Honest limits

JevBench, the board's open-weights rankingDot and interval chart of the board's open-weights ranking, the first ten of 128 with Jev 1.13.0 as a reference row, on JevBench v1.6.1; every row is in the table on this page and on the benchmarks page.646668707274761. H2O-Lightning-4B v1.172.52 [69.79, 73.79]2. decisio v0.8.0, Gemma 4 31B (ours)71.69 [69.86, 72.81]3. Quyet-1.0-Large71.39 [69.84, 72.21]4. decider-12b v270.88 [68.54, 72.21]5. decider-12b v170.42 [68.09, 71.63]6. torchcast-decision-12b69.91 [67.22, 71.47]7. Winnow-12B Q868.89 [66.45, 70.46]8. deck-31B68.68 [67.07, 69.74]9. Cygnet68.55 [64.42, 69.94]10. decisio v0.8.0, Gemma 4 12B (ours)68.17 [57.37, 69.65]Jev 1.13.071.49 [69.14, 72.40]JevBench score, with its 95% interval
JevBench: the board's open-weights ranking, first ten of 128, with Jev as a reference row
Open-weights rankSystemScore95% interval
1H2O-Lightning-4B v1.172.5269.79 to 73.79
2decisio v0.8.0, Gemma 4 31B (ours)71.6969.86 to 72.81
3Quyet-1.0-Large71.3969.84 to 72.21
4decider-12b v270.8868.54 to 72.21
5decider-12b v170.4268.09 to 71.63
6torchcast-decision-12b69.9167.22 to 71.47
7Winnow-12B Q868.8966.45 to 70.46
8deck-31B68.6867.07 to 69.74
9Cygnet68.5564.42 to 69.94
10decisio v0.8.0, Gemma 4 12B (ours)68.1757.37 to 69.65
referenceJev 1.13.071.4969.14 to 72.40
The JevBench board: its file for revision v1.6.1, sha256 5cd8c1332226.

How it works

Read once. Answer fast.

No training: one pass for each question.

One request carries a state and its questions.

The requestYour app sends one request with a state and its questions. Decisio answers each question in one pass, and the response is a probability for every option; the bars are a recorded answer.Your appa state, questionsPOST /v1/systemoneone requestDecisioone pass a questiona probability for every optionanswer0.96search0.01calculate0.02ask_user0.00act0.01

The state is read once. Every later question starts from the cache.

The cacheThe first question about a new 3,000-token document takes 490 milliseconds because it reads the document. Each later question takes 32 to 43 ms. The bars are to scale.One 3,000-token document, Decisio 31B, to scalefirst question: 490 ms32 to 43 ms eachreads the documenttime

Keep the answers above a threshold you pick. Send the rest to a person or a larger model.

The thresholdA bar of the 1,172 questions of example 5: the ones kept at or above the threshold, and the ones escalated to a person or a larger model. The threshold moves through the measured values.Keep answers at or above0.827 of 1,172 questions escalatedThe rest are kept. Drawn to scale.

Speed, measured on one box

Many questions about one document

Fifty questions about one document, on one box.

  • about 44 timesas long for Jeff and Quyet as for Decisio 31B, at 3,000 tokens, on the same box
  • 66 timesas long for them at 6,000 tokens
  • 10.1 sfor Decisio 31B to answer 50 questions about a 32,000-token document, which neither Jeff nor Quyet accepts
Bar chart of the median time to answer 50 questions about one document, five systems measured in one session on one box. For a 3,000-token document: decisio Gemma 4 31B 1.08 seconds, decisio Gemma 4 12B 1.57, H2O-Lightning-4B 2.05, Jeff-1.0-Large 47.1, Quyet-1.0-Large 48.9. For a 6,000-token document: 1.73, 1.97, 2.82, 114.6 and 114.8 seconds. For a 32,000-token document only decisio Gemma 4 12B (6.76 s), H2O-Lightning-4B (9.05 s) and decisio Gemma 4 31B (10.1 s) answer; Jeff-1.0-Large and Quyet-1.0-Large are marked not supported.Bar chart of the median time to answer 50 questions about one document, five systems measured in one session on one box. For a 3,000-token document: decisio Gemma 4 31B 1.08 seconds, decisio Gemma 4 12B 1.57, H2O-Lightning-4B 2.05, Jeff-1.0-Large 47.1, Quyet-1.0-Large 48.9. For a 6,000-token document: 1.73, 1.97, 2.82, 114.6 and 114.8 seconds. For a 32,000-token document only decisio Gemma 4 12B (6.76 s), H2O-Lightning-4B (9.05 s) and decisio Gemma 4 31B (10.1 s) answer; Jeff-1.0-Large and Quyet-1.0-Large are marked not supported.
The exact wording, from the facts

On one box, Decisio 31B answers 50 questions about one 3,000-token document in 1.08 s and about one 6,000-token document in 1.73 s; Decisio 12B takes 1.57 s and 1.97 s, and H2O-Lightning-4B 2.05 s and 2.82 s.

On one box, Jeff-1.0-Large and Quyet-1.0-Large, the two systems ranked above Decisio on JevBench's capability score, take about 44 times (3,000 tokens) and 66 times (6,000 tokens) as long as Decisio 31B to answer 50 questions about one document, and neither accepts a 32,000-token document.

At 32,000 tokens, Decisio 31B answers 50 questions about one document in 10.1 s, Decisio 12B in 6.76 s and H2O-Lightning-4B in 9.05 s; Jeff-1.0-Large accepts documents of up to 8,192 tokens and Quyet-1.0-Large up to 6,000, so neither accepts a 32,000-token document.

On one box, the first question about a new 3,000-token document takes a median of 112 ms on H2O-Lightning-4B, 295 ms on Decisio 12B, 490 ms on Decisio 31B, and 895 ms on Jeff-1.0-Large and 895 ms on Quyet-1.0-Large.

The board, column by column

JevBench, axis by axis

The ten highest open-weights systems and Jev on the public board, each cell shaded by its rank in the column; higher is better on every axis.

Table of JevBench v1.6.1 scores, 11 rows sorted by composite score: the ten highest open-weights systems and Jev 1.13.0 as a reference row, six columns (overall, intelligence, sealed questions, calibration, speed, cost), each cell shaded by its rank in the column. Decisio 31B v0.8.0's values are overall 71.69, intelligence 70.4, sealed questions 73.3, calibration 88.7, speed 91.2, cost 51.7. Among the 11 rows shown it is second on overall, third on intelligence, second on sealed questions, fifth on calibration, fifth on speed and ninth on cost. The highest in each column: overall: H2O-Lightning-4B v1.1; intelligence: Quyet-1.0-Large; sealed questions: deck-31B; calibration: Jev 1.13.0; speed: H2O-Lightning-4B v1.1; cost: H2O-Lightning-4B v1.1. The Decisio rows are the board's v0.8.0 entries: Decisio 31B v0.8.0 (71.69) and Decisio 12B v0.8.0 (68.17). The board's decisio v0.9.0 12B entry (67.76) is not among the rows shown.Table of JevBench v1.6.1 scores, 11 rows sorted by composite score: the ten highest open-weights systems and Jev 1.13.0 as a reference row, six columns (overall, intelligence, sealed questions, calibration, speed, cost), each cell shaded by its rank in the column. Decisio 31B v0.8.0's values are overall 71.69, intelligence 70.4, sealed questions 73.3, calibration 88.7, speed 91.2, cost 51.7. Among the 11 rows shown it is second on overall, third on intelligence, second on sealed questions, fifth on calibration, fifth on speed and ninth on cost. The highest in each column: overall: H2O-Lightning-4B v1.1; intelligence: Quyet-1.0-Large; sealed questions: deck-31B; calibration: Jev 1.13.0; speed: H2O-Lightning-4B v1.1; cost: H2O-Lightning-4B v1.1. The Decisio rows are the board's v0.8.0 entries: Decisio 31B v0.8.0 (71.69) and Decisio 12B v0.8.0 (68.17). The board's decisio v0.9.0 12B entry (67.76) is not among the rows shown.
  • On JevBench v1.6.1, among the ten highest open-weights systems and Jev: 2nd overall, 2nd on sealed questions (73.3, never published) and 3rd on intelligence (70.4 against Jev's 63.6).
  • On JevBench v1.6.1, within 2 points of the highest on calibration (88.7; Jev 90.6) and on speed (91.2; H2O-Lightning-4B 92.6).
  • On JevBench v1.6.1, 9th on the board's cost axis (51.7; H2O-Lightning-4B 60.3 is highest). On your own hardware you pay for the card, not per token.
Every number
JevBench: the ten highest open-weights systems and Jev 1.13.0 on the composite, the four axes and the sealed questions, with each value's rank in its column
SystemOverallIntelligenceSealed questionsCalibrationSpeedCost
H2O-Lightning-4B72.5260.059.090.092.660.3
Decisio 31B v0.8.0 (ours)71.6970.473.388.791.251.7
Jev 1.13.0 (reference)71.4963.661.690.691.554.7
Quyet-1.0-Large71.3973.471.690.086.950.5
decider-12b v270.8863.063.682.090.557.8
decider-12b v170.4260.659.583.890.557.8
torchcast-decision-12b69.9160.561.182.991.756.4
Winnow-12B Q868.8959.559.283.086.756.6
deck-31B68.6873.075.882.286.749.7
Cygnet68.5554.851.387.091.856.4
Decisio 12B v0.8.0 (ours)68.1752.351.689.287.059.3

The highest value in each column is in bold. Source: the board's file for revision v1.6.1.

Routing

Use the probability

Keep the answers above a threshold you pick and send the rest to a person or a larger model. Example 5 measured what that buys.

Every threshold, as measured
RuleEscalatedRight on the rest95% intervalConfidently wrong (wrong, kept)Wrong answers escalated
No escalation0 (0.0%)97.4%not applicable310 of 31
Keep at 0.8 or more27 (2.3%)98.5%97.6% to 99.1%1714 of 31
Keep at 0.9 or more47 (4.0%)98.8%98.0% to 99.3%1318 of 31
Line chart of accuracy on the questions kept against the share of questions escalated, for thresholds from 0.5 to 0.95 on Gemma 4 31B, with the 0.8 and 0.9 points marked. Example 5, Escalate only when unsure: ARC-Challenge test split, a public benchmark the model may have seen; decisio 0.9.0 on Gemma 4 31B, one NVIDIA RTX PRO 6000 Blackwell at 600 W, recorded 2026-10-08.Line chart of accuracy on the questions kept against the share of questions escalated, for thresholds from 0.5 to 0.95 on Gemma 4 31B, with the 0.8 and 0.9 points marked. Example 5, Escalate only when unsure: ARC-Challenge test split, a public benchmark the model may have seen; decisio 0.9.0 on Gemma 4 31B, one NVIDIA RTX PRO 6000 Blackwell at 600 W, recorded 2026-10-08.

Why

Why Decisio

  • Many questions, one read

    Every question in a request is scored from one read of your text: the same choice as one question at a time on every item measured, probabilities within 0.0073.

  • Long prompts, many options

    Prompts up to 32,768 tokens and 255 options per question.

  • Your hardware, no per-token bill

    It runs on your own GPU, Mac or laptop; the cost is the card, not a bill per token.

Demos

Watch it decide

Decisions in real time, from Decisio's own demos and one recorded run. Each clip is cut at the speed it happened. Nothing is sped up.

The driving clip draws map data from OpenStreetMap (ODbL, © OpenStreetMap contributors).

Get started

Four ways in, one command each.

# the Gemma 4 12B base, for 16 GB laptopsollama pull aminroudaki/decisio-gemma# the Qwen baseollama pull aminroudaki/decisio

The Ollama guide, and every path.

git clone https://github.com/aminry/decisiocd decisiouv sync --extra serve --frozenuv run python -m decisio.serve.vllm_engine

The GPU guide, and every path.

uv build --wheel  # the image installs this wheeldocker compose up --build  # needs the NVIDIA Container Toolkitcurl http://127.0.0.1:8000/health  # answers once the first start has fetched the checkpoint

The Docker guide, and every path.

uv sync --extra mlxuv run python -m decisio.serve.vllm_engine --backend mlx --model mlx-community/Qwen3.6-35B-A3B-6bituv run python -m decisio.serve.vllm_engine --backend mlx --base gemma-4-12b --model mlx-community/gemma-4-12B-it-6bituv run python -m decisio.serve.vllm_engine --backend mlx --base gemma-4-31b --model mlx-community/gemma-4-31b-it-6bit

The MLX guide, and every path.

Read this

Honest limits

  • One box, one sessionEvery race figure is from one RTX PRO 6000 Blackwell Workstation Edition, power limit 600 W (default, enforced), each system on its own server.
  • One questionThe one-question figure in the hero is server time, on decisio 0.8.1's served defaults and one RTX PRO 6000 at 585 W, so a card at 600 W may read new states faster.
  • Time, not accuracyThe race compares how long fifty questions take, on synthetic support tickets.
  • Replayed, not liveThe race replays medians of 5 or 10 documents, and the demo replays decisions recorded on decisio 0.9.0.
  • A snapshot of the boardThe ranks are for JevBench v1.6.1; the live board has moved since.
  • HardwareGemma 4 31B needs a 96 GB card, or a 64 GB Mac.

Ask it something real.

ollama pull aminroudaki/decisio-gemma
Read the quickstartStar it on GitHub