Metadata-Version: 2.5
Name: jev-lite
Version: 0.0.7
Summary: OpenAI-style sync and async Pydantic client for TypeSafe Jev (System One).
Project-URL: Documentation, https://docs.typesafe.ai/
Project-URL: Local model, https://huggingface.co/dockhardman/d3-nano-onnx
Author-email: Allen Chou <f1470891079@gmail.com>
License-Expression: Apache-2.0
License-File: LICENSE
Keywords: jev,openai,opentelemetry,pydantic,system-one,typesafe
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Classifier: Programming Language :: Python :: 3.14
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Classifier: Topic :: Software Development :: Libraries :: Python Modules
Classifier: Typing :: Typed
Requires-Python: >=3.11
Requires-Dist: httpx>=0.23
Requires-Dist: opentelemetry-api>=1.20
Requires-Dist: opentelemetry-exporter-otlp-proto-http>=1.20
Requires-Dist: opentelemetry-sdk>=1.20
Requires-Dist: pydantic>=2
Provides-Extra: cache
Requires-Dist: diskcache>=5.6; extra == 'cache'
Provides-Extra: distill
Requires-Dist: datasets>=2; extra == 'distill'
Provides-Extra: local
Requires-Dist: numpy>=1.24; extra == 'local'
Requires-Dist: onnxruntime>=1.24; extra == 'local'
Requires-Dist: tokenizers>=0.15; extra == 'local'
Provides-Extra: ml
Requires-Dist: accelerate>=1.0; extra == 'ml'
Requires-Dist: pillow>=9; extra == 'ml'
Requires-Dist: torch>=2.1; extra == 'ml'
Requires-Dist: torchvision>=0.16; extra == 'ml'
Requires-Dist: transformers>=4.40; extra == 'ml'
Description-Content-Type: text/markdown

# jev-lite

OpenAI-style Python client for [TypeSafe Jev](https://typesafe.ai/) (System One).

Jev does not generate text. Send `state` plus typed questions; get structured
decisions with probabilities. `jev-lite` is the official TypeSafe HTTP client — sync
and async, same shape as the OpenAI Python SDK. `jev-lite@0.0.2` added Dialogflow-style
system entities; `jev-lite@0.0.3` adds a privacy filter and a reranker (all
three are previews, below); `jev-lite@0.0.4` adds images and Box answers
for Jev 1.14; `jev-lite@0.0.5` adds `LocalTypeSafe`, the same API answered by
an open model on your CPU with no API key; `jev-lite@0.0.6` adds
`OpenAIDecisions`, the same API answered by OpenAI's Decisions API;
`jev-lite@0.0.7` makes `LocalTypeSafe` read the state once per request, and
measures [latency by question count](#latency-by-question-count) for every
model on the board.

```bash
pip install jev-lite
# or: uv add jev-lite
```

Python 3.11+. Set `TYPESAFE_API_KEY`. Docs: <https://docs.typesafe.ai/>

## Usage

```python
from jev_lite import Choice, Noul, Score, TypeSafe

client = TypeSafe()  # TYPESAFE_API_KEY

result = client.system_one.create(
    state="I was charged twice. Please fix this ASAP.",
    questions={
        "billing": Noul(instructions="Is this ticket about billing?"),
        "tone": Choice(
            instructions="What is the customer's tone?",
            criteria={"calm": None, "frustrated": None, "angry": None},
        ),
        "urgency": Score(
            instructions="How urgent is this ticket?",
            criteria=["can wait", "this week", "today"],
        ),
    },
)

print(result.nouls["billing"].noul)
print(result.choices["tone"].choice)
print(result.scores["urgency"].score)
print(result.elapsed_ms, result.request_id, result.usage)
print(client.models.list())  # GET /v1/models
```

Async is the same surface with `await`:

```python
from jev_lite import AsyncTypeSafe, Noul

async with AsyncTypeSafe() as client:
    result = await client.system_one.create(
        state="I was charged twice. Please fix this ASAP.",
        questions={"billing": Noul(instructions="Is this ticket about billing?")},
    )
    models = await client.models.list()
```

`jev-latest` is an alias that moves: on 2026-10-21 it answers as
`jev-1.13.0`, and `jev-1.14.0` is opt-in. `client.models.list()` lists only
the aliases (`jev-latest`, `jev-preview`); `response.model` names the version
that answered. Pin a version with `TypeSafe(model="jev-1.14.0")` (or
`TYPESAFE_DEFAULT_MODEL`) when answers must repeat. Prefer one
`system_one.create` with many questions over a loop of calls. From 0.0.7 the
modules are pinned to `jev-1.14.0` and use the client's `model`. On the same
day, 1.14 matched or beat 1.13 on every module below, and scored lower on
hard semantic decisions (461 vs 481 of 504 on `semantic-decisions@v0.0.2`).

Re-running unchanged questions can replay answers from disk
(`pip install "jev-lite[cache]"`):

```python
client = TypeSafe(model="jev-1.14.0", cache="~/.cache/jev")  # or a diskcache.Cache
```

Only 200 responses are stored, keyed on the request body (which carries the
model id). Pin a version: an alias such as `jev-latest` can move to a new model
while the cache keeps replaying the old one, so a cached alias warns. Leave the
cache off when you measure repeat or order stability.

## Local decisions (CPU, no API key)

`LocalTypeSafe` answers the same `system_one.create` calls on this machine's
CPU with [`vllm-sr/d3-nano`](https://huggingface.co/vllm-sr/d3-nano) (2.2B,
Apache-2.0), exported to ONNX Runtime with int8 weights. No torch, no
transformers, no remote code.

```bash
pip install "jev-lite[local]"   # about 40 MB download, 150 MB installed
```

```python
from jev_lite import Choice, LocalTypeSafe

client = LocalTypeSafe()  # model "vllm-sr/d3-nano"
client.load()             # optional warm-up; the first run downloads 1.76 GiB
result = client.system_one.create(
    state="包裹顯示已送達，可是我家門口什麼都沒有。",
    questions={"route": Choice(criteria={"billing": "付款", "shipping": "配送"})},
)
print(result.choices["route"].choice)  # shipping
```

- **Weights:** [`dockhardman/d3-nano-onnx`](https://huggingface.co/dockhardman/d3-nano-onnx)
  at a pinned commit, each file checked against its sha256 and cached in
  `~/.cache/jev-lite` (`cache_dir=`); after that it runs offline.
- **Same API:** `AsyncLocalTypeSafe`, `models.list()`, `cache=` and the
  tracing spans work as for `TypeSafe`. Text only: an image or a `Box`
  question raises `BadRequestError`; a question over 8,192 tokens is
  refused, never truncated.
- **Quality:** 449/504 on the zh-Hant `semantic-decisions@v0.0.2` (hard 296/336), against
  481 for hosted `jev-1.13.0`, with the same choice as the original bf16
  model on all 504. The modules below are measured on hosted Jev only.
- **Speed:** one question at a time on 20 Arm cores: a median of 0.5 s on
  short questions and 1.1 s on the bank's prompts, about 5 s at 1,100 tokens.

## OpenAI Decisions

`OpenAIDecisions` sends the same `system_one.create` calls to OpenAI's
Decisions API (`POST /v1/decisions`). It needs no extra install; only the
client class changes.

```python
from jev_lite import Choice, OpenAIDecisions

client = OpenAIDecisions()  # OPENAI_API_KEY; model "gpt-6-luna"
result = client.system_one.create(
    state="包裹顯示已送達，可是我家門口什麼都沒有。",
    questions={"route": Choice(criteria={"billing": "付款", "shipping": "配送"})},
)
print(result.choices["route"].choice)  # shipping
```

- **Mapping:** `Noul` → predicate (its `true`/`false` texts are appended to
  the instructions), `Choice` → choice, and `Score` → score levels. A
  question without instructions gets a short default. `state` that is not a
  string is sent as compact JSON text.
- **Images:** an `Image` in `state` or `instructions` is sent as an image
  part. An image in `criteria` and a `Box` question raise `BadRequestError`
  before anything is sent, as do counts the API refuses (1–400 questions,
  2–255 options, 2–10 levels).
- **Same API:** `AsyncOpenAIDecisions`, `cache=`, typed errors, `request_id`,
  `usage` and the tracing spans work as for `TypeSafe`. `models.list()`
  returns the `luna` models.
- **Quality:** `gpt-6-luna` scored 442/504 on the zh-Hant
  `semantic-decisions@v0.0.2` (hard 275/336), against 481 for Jev
  `jev-1.13.0` and 449 for `LocalTypeSafe`. It changed its answer with option
  order on 20 of 252 rows.

## Images and Box (Jev 1.14, experimental)

Put an `Image` anywhere a value can go: in `state`, `instructions` or
`criteria`. The client uploads it as an asset and adds the
`TypeSafe-Danger-Zone-Images` header. A `Box` answer is where a target is,
as edges in fractions of the image; it needs `jev-1.14.0`.

```python
from pathlib import Path
from jev_lite import Box, Image, Noul, TypeSafe

with TypeSafe(model="jev-1.14.0") as client:
    result = client.system_one.create(
        state={"photo": Image(Path("kitchen.jpg"))},  # PNG/JPEG, sides 280-1568 px
        questions={
            "has_mug": Noul(instructions="Is there a coffee mug in `photo`?"),
            "mug": Box(instructions="the coffee mug in `photo`"),
        },
    )

mug = result.boxes["mug"]
print(mug.box.left, mug.box.top, mug.box.right, mug.box.bottom, mug.confidence)
```

A Box always returns a box, so gate it on the Noul. An image costs about
`2·⌈w/28⌉·⌈h/28⌉` input tokens. Measured limits (targets under ~40 px,
counting past five, text under 7 px); the benchmark is
`image-questions@v0.0.1`.

## System entities (preview)

`jev_lite.system_entities` extracts the 59 Dialogflow ES system entities
(`@sys.date-time`, `@sys.unit-currency`, `@sys.geo-city`, …) and fills
typed slots, in ten languages (en and zh-Hant primary). Code proposes
candidates, Jev decides (the client's `model`), code normalizes. **Preview:** the API
may change within `0.0.x`. Same install; no extra dependency.

```python
from datetime import datetime
from zoneinfo import ZoneInfo

from jev_lite import TypeSafe
from jev_lite.system_entities import extract

now = datetime(2026, 10, 1, 16, 0, tzinfo=ZoneInfo("Asia/Taipei"))
with TypeSafe() as client:
    result = extract(client, "Book a table for 4 next Friday at 7pm", ["date-time", "number"], now=now)
for match in result.matches:
    print(match.entity, match.span.text, match.value)  # number 4 …; date-time … 2026-10-09 19:00
```

Slot filling (`extract_slots`), async (`aextract`, `aextract_slots`) and the
value models are documented in the module's docstrings. Extraction is not a
security boundary, `probability` is not calibrated, and th, ms and other
languages outside the ten get few matches. On the held-out set
`system-entities-cases@v0.0.1` (301 cases, ten languages) `jev-lite@0.0.5` (unchanged in 0.0.6) is
98.9% exact on accepted values with 98.9% coverage on `jev-1.13.0`
(`jev-lite@0.0.4`: 98.9% on `jev-1.13.0`, 99.3% on `jev-1.14.0`).

## Privacy filter and reranker (preview)

`jev_lite.privacy_filter` and `jev_lite.reranker` ship in `jev-lite@0.0.3`
as previews: the APIs may change within `0.0.x`. Import them as modules
(`from jev_lite.privacy_filter import detect`, `from jev_lite.reranker import
rerank`). Both use the client's `model` and are pinned to `jev-1.14.0` from
0.0.7; the numbers are on `jev-1.13.0`, with the same-day 1.14 run beside them.

**Privacy filter**: private-span detection. Public multilingual gold
(ai4privacy OpenPII, Gretel finance, NVIDIA Nemotron-PII), 300 test texts,
exact-span precision / recall:

| Setup | P / R | Over-redaction | Speed |
| --- | --- | --- | --- |
| [openai/privacy-filter](https://huggingface.co/openai/privacy-filter), local GPU | 63.4 / 62.9 | 5.2% | ~20 ms per text |
| openai/privacy-filter proposes, Jev decides | 75.4 / 60.1 | 2.5% | p50 210 ms |
| Jev only, no GPU (broader proposals, per-label cuts) | 68.6 / 60.5 | 2.4% | p50 215 ms |
| Jev only, round-04 default | 80.4 / 57.6 | 1.5% | p50 210 ms |
| Jev only, before the person proposer | 78.3 / 63.6 | 1.6% | p50 209 ms |
| **Jev only, default `detect(client, text)`**, no GPU | **79.3 / 69.2** | **1.6%** | p50 ~0.7 s |
| Same default on `jev-1.14.0` (same-day 1.13: 79.7 / 69.7, 1.6%) | 82.4 / 68.2 | 1.3% | call p50 0.19 s |

The default uses per-label proposers and cuts fitted on dev, and finds
people with `system_entities` person spans plus a person detect Noul (person
recall 43.9 → 71.9), with an order-stable name boundary. On overlap it
scores 88.9 / 77.6; exact address recall is 0 because the gold splits an
address into components while the filter masks the whole address. It fails
most written robustness bars, so it is a preview. Hosted Jev sees the raw
text before masking.

**Reranker**: batched evidence Noul over a first-stage pool (balanced
batches of at most 30, independent of the caller's order), nDCG@10:

| Bank | First stage | Reranked | Query p50 | Same day, 1.13 → 1.14 |
| --- | --- | --- | --- | --- |
| SciFact (50 q) | .556 | .638 | 0.30 s | .635 → .634 |
| FiQA (50 q) | .219 | .346 | 0.26 s | .346 → .348 |
| zh, BM25 (50 q) | .526 | .587 | 0.30 s | .603 → .618 |
| zh, E5 (50 q) | .551 | .647 | 0.27 s | .640 → .656 |
| Mr. TyDi ko / ja / ar (30 q each) | .415 / .605 / .529 | .675 / .791 / .754 | 0.23–0.24 s | .687 / .791 / .741 → .675 / .808 / .741 |

A batch score is relative to the other passages in its request, not a
standalone probability. A passage that echoes the relevance criteria can be
lifted into the top 10; `rerank(..., check_echo=True)` (off by default)
re-checks the top-k and moves such passages down, at about +0.23 s.

## Tracing

Off by default. Opt in and pick the destination:

```python
from jev_lite import instrument

instrument(endpoint="http://127.0.0.1:4318")  # local Grafana LGTM
```

Or set `OTEL_EXPORTER_OTLP_ENDPOINT` and call `instrument()`, or install your
own `TracerProvider` first (Logfire, etc.). Spans record prompt, result, and
latency.

## Upgrading from 0.0.1–0.0.5

- **Pin the model.** `jev-latest` moves to new versions, and answers and
  confidence shift with it. Use `TypeSafe(model="jev-1.14.0")` (0.0.7; it was
  `jev-1.13.0` before), and refit any confidence cut when you move.
  A cached client asking for an alias warns (0.0.6).
- **`Noul(true=..., false=...)`** works from 0.0.5. Before that, the
  shorthand was silently ignored; use `criteria=NoulCriteria(...)` there.
  Unknown question fields warn from 0.0.5.
- **Replay cache:** `TypeSafe(cache="dir")` (0.0.3, `jev-lite[cache]`)
  replaces hand-made caches. `jev_lite.eval` is research code, not a stable
  API.
- **Confidence** comes with every Choice and Score answer; do not recompute
  it. From 0.0.6, `TypeSafe`, `LocalTypeSafe` and `OpenAIDecisions` all
  report `(n·p_max − 1)/(n − 1)`.
- **Other backends behind the same calls:** `LocalTypeSafe` (0.0.5, CPU, no
  key) and `OpenAIDecisions` (0.0.6). A self-hosted server that speaks
  `/v1/systemone` works through `TypeSafe(base_url=...)`.
- **Modules** (`system_entities`, `privacy_filter`, `reranker`) use the
  client's model from 0.0.5. From 0.0.7 they are pinned to `jev-1.14.0`;
  their tables give 1.13 and 1.14 from the same day.

## Develop

```bash
uv sync --dev
uv run ruff check . && uv run pytest
```

The supported public API is the TypeSafe client. Optional local engines live
under `jev_lite.engine` (`pip install 'jev-lite[ml]'`).

## Semantic decisions v0.0.2

Dataset `semantic-decisions@v0.0.2`, versioned separately from the package.

252 frozen zh-Hant rows
in 18 families (36 easy, 48 medium, 168 hard), gold derived from visible text.
This natural zh-Hant v1 qualified freeze **is** v0.0.2, and it is the only
semantic-decision eval in this repo. Every model answers the same 504 locked
requests: each row in canonical and left-rotated option order. Charts are
percent correct. The families were seen during development before the
freeze, so this is not unseen confirmation.

[![All 504](https://quickchart.io/chart?version=3&w=640&h=1210&bkg=white&c=%7B%22type%22%3A%22bar%22%2C%22data%22%3A%7B%22labels%22%3A%5B%22d3%2027B%22%2C%22AutoTrust%20v1.1%22%2C%22AutoTrust%20JEV-27B%22%2C%22AutoTrust%20JEV-27B-VL%22%2C%22pplx%2027B%20v1.1%22%2C%22Jev%201.13.0%22%2C%22AutoTrust%20JEV-35B%22%2C%22d3-flash%209B%22%2C%22linne-oak%2031B%22%2C%22StartLux%2027B%22%2C%22d3-mini%204B%22%2C%22Shisa%20DE-2%22%2C%22Jeff-1.0-Large%22%2C%22AutoJev%2027B%22%2C%22Rune%20v3%2026B-A4B%22%2C%22decider%2031B%22%2C%22Jev%201.14.0%22%2C%22pplx%20flash%204B%22%2C%22d3-nano%202B%22%2C%22Nimble%209B%20v3%22%2C%22OpenAI%20gpt-6-luna%22%2C%22h2o%20lightning%204B%22%2C%22decider-4b%20v2%22%2C%22NeoHorse-Jev%204B%22%2C%22Tev1%204B%22%2C%22AutoTrust%20JEV-9B%22%2C%22Clef%2027B%22%2C%22Mica%20v0.1%204B%22%2C%22Kev%204B%22%2C%22Clef-Flash%209B%22%2C%22RSI-Jev%20v4.0-VL%22%2C%22StandardOne%203B%22%2C%22RSI-Jev%20v3.0%22%2C%22d3-lite%200.8B%22%2C%22Strands%202B%22%2C%22Tev1%200.8B%22%2C%22semif%22%2C%22JevEmbed%204B%22%2C%22Lev%204B%22%2C%22decider%200.8B%22%2C%22JevK5%20v0.2%22%2C%22Plumb%204B%22%2C%22GLiNER%22%2C%22Laya%22%2C%22jeff%22%5D%2C%22datasets%22%3A%5B%7B%22data%22%3A%5B98.8%2C98.4%2C96.4%2C96.4%2C96.2%2C95.4%2C95.4%2C95.2%2C95.2%2C95.2%2C94.6%2C93.7%2C93.5%2C93.3%2C92.9%2C92.5%2C91.5%2C89.5%2C89.1%2C89.1%2C87.7%2C86.7%2C86.5%2C84.9%2C84.3%2C82.5%2C80.2%2C80.0%2C79.4%2C76.2%2C74.6%2C72.8%2C72.4%2C70.0%2C69.8%2C66.3%2C65.9%2C65.5%2C62.9%2C62.1%2C57.5%2C55.8%2C51.0%2C47.6%2C43.5%5D%2C%22backgroundColor%22%3A%5B%22%23b91c1c%22%2C%22%23e879f9%22%2C%22%23c026d3%22%2C%22%239333ea%22%2C%22%23334155%22%2C%22%230d9488%22%2C%22%237e22ce%22%2C%22%23ef4444%22%2C%22%234d7c0f%22%2C%22%231d4ed8%22%2C%22%23f97316%22%2C%22%2306b6d4%22%2C%22%23be185d%22%2C%22%23d97706%22%2C%22%23475569%22%2C%22%23115e59%22%2C%22%2314b8a6%22%2C%22%2364748b%22%2C%22%23f59e0b%22%2C%22%230e7490%22%2C%22%2310a37f%22%2C%22%23facc15%22%2C%22%230f766e%22%2C%22%232563eb%22%2C%22%23ea580c%22%2C%22%23a21caf%22%2C%22%23059669%22%2C%22%23dc2626%22%2C%22%2392400e%22%2C%22%2310b981%22%2C%22%230ea5e9%22%2C%22%238b5cf6%22%2C%22%2338bdf8%22%2C%22%23fdba74%22%2C%22%2384cc16%22%2C%22%230891b2%22%2C%22%234f46e5%22%2C%22%23a16207%22%2C%22%237c3aed%22%2C%22%2316a34a%22%2C%22%230369a1%22%2C%22%23b45309%22%2C%22%2365a30d%22%2C%22%23e11d48%22%2C%22%23db2777%22%5D%2C%22borderWidth%22%3A0%2C%22barPercentage%22%3A0.82%7D%5D%7D%2C%22options%22%3A%7B%22indexAxis%22%3A%22y%22%2C%22layout%22%3A%7B%22padding%22%3A%7B%22right%22%3A40%7D%7D%2C%22plugins%22%3A%7B%22legend%22%3A%7B%22display%22%3Afalse%7D%2C%22title%22%3A%7B%22display%22%3Atrue%2C%22text%22%3A%22All%20504%22%2C%22font%22%3A%7B%22size%22%3A16%7D%7D%2C%22datalabels%22%3A%7B%22display%22%3Atrue%2C%22anchor%22%3A%22end%22%2C%22align%22%3A%22end%22%2C%22clamp%22%3Atrue%2C%22color%22%3A%22%23111827%22%2C%22font%22%3A%7B%22size%22%3A11%7D%2C%22formatter%22%3Afunction%28v%29%7Breturn%20v.toFixed%281%29%7D%7D%7D%2C%22scales%22%3A%7B%22x%22%3A%7B%22min%22%3A0%2C%22max%22%3A100%2C%22ticks%22%3A%7B%22stepSize%22%3A20%7D%2C%22title%22%3A%7B%22display%22%3Atrue%2C%22text%22%3A%22%25%22%7D%7D%2C%22y%22%3A%7B%22grid%22%3A%7B%22display%22%3Afalse%7D%7D%7D%7D%7D)](https://quickchart.io/chart?version=3&w=640&h=1210&bkg=white&c=%7B%22type%22%3A%22bar%22%2C%22data%22%3A%7B%22labels%22%3A%5B%22d3%2027B%22%2C%22AutoTrust%20v1.1%22%2C%22AutoTrust%20JEV-27B%22%2C%22AutoTrust%20JEV-27B-VL%22%2C%22pplx%2027B%20v1.1%22%2C%22Jev%201.13.0%22%2C%22AutoTrust%20JEV-35B%22%2C%22d3-flash%209B%22%2C%22linne-oak%2031B%22%2C%22StartLux%2027B%22%2C%22d3-mini%204B%22%2C%22Shisa%20DE-2%22%2C%22Jeff-1.0-Large%22%2C%22AutoJev%2027B%22%2C%22Rune%20v3%2026B-A4B%22%2C%22decider%2031B%22%2C%22Jev%201.14.0%22%2C%22pplx%20flash%204B%22%2C%22d3-nano%202B%22%2C%22Nimble%209B%20v3%22%2C%22OpenAI%20gpt-6-luna%22%2C%22h2o%20lightning%204B%22%2C%22decider-4b%20v2%22%2C%22NeoHorse-Jev%204B%22%2C%22Tev1%204B%22%2C%22AutoTrust%20JEV-9B%22%2C%22Clef%2027B%22%2C%22Mica%20v0.1%204B%22%2C%22Kev%204B%22%2C%22Clef-Flash%209B%22%2C%22RSI-Jev%20v4.0-VL%22%2C%22StandardOne%203B%22%2C%22RSI-Jev%20v3.0%22%2C%22d3-lite%200.8B%22%2C%22Strands%202B%22%2C%22Tev1%200.8B%22%2C%22semif%22%2C%22JevEmbed%204B%22%2C%22Lev%204B%22%2C%22decider%200.8B%22%2C%22JevK5%20v0.2%22%2C%22Plumb%204B%22%2C%22GLiNER%22%2C%22Laya%22%2C%22jeff%22%5D%2C%22datasets%22%3A%5B%7B%22data%22%3A%5B98.8%2C98.4%2C96.4%2C96.4%2C96.2%2C95.4%2C95.4%2C95.2%2C95.2%2C95.2%2C94.6%2C93.7%2C93.5%2C93.3%2C92.9%2C92.5%2C91.5%2C89.5%2C89.1%2C89.1%2C87.7%2C86.7%2C86.5%2C84.9%2C84.3%2C82.5%2C80.2%2C80.0%2C79.4%2C76.2%2C74.6%2C72.8%2C72.4%2C70.0%2C69.8%2C66.3%2C65.9%2C65.5%2C62.9%2C62.1%2C57.5%2C55.8%2C51.0%2C47.6%2C43.5%5D%2C%22backgroundColor%22%3A%5B%22%23b91c1c%22%2C%22%23e879f9%22%2C%22%23c026d3%22%2C%22%239333ea%22%2C%22%23334155%22%2C%22%230d9488%22%2C%22%237e22ce%22%2C%22%23ef4444%22%2C%22%234d7c0f%22%2C%22%231d4ed8%22%2C%22%23f97316%22%2C%22%2306b6d4%22%2C%22%23be185d%22%2C%22%23d97706%22%2C%22%23475569%22%2C%22%23115e59%22%2C%22%2314b8a6%22%2C%22%2364748b%22%2C%22%23f59e0b%22%2C%22%230e7490%22%2C%22%2310a37f%22%2C%22%23facc15%22%2C%22%230f766e%22%2C%22%232563eb%22%2C%22%23ea580c%22%2C%22%23a21caf%22%2C%22%23059669%22%2C%22%23dc2626%22%2C%22%2392400e%22%2C%22%2310b981%22%2C%22%230ea5e9%22%2C%22%238b5cf6%22%2C%22%2338bdf8%22%2C%22%23fdba74%22%2C%22%2384cc16%22%2C%22%230891b2%22%2C%22%234f46e5%22%2C%22%23a16207%22%2C%22%237c3aed%22%2C%22%2316a34a%22%2C%22%230369a1%22%2C%22%23b45309%22%2C%22%2365a30d%22%2C%22%23e11d48%22%2C%22%23db2777%22%5D%2C%22borderWidth%22%3A0%2C%22barPercentage%22%3A0.82%7D%5D%7D%2C%22options%22%3A%7B%22indexAxis%22%3A%22y%22%2C%22layout%22%3A%7B%22padding%22%3A%7B%22right%22%3A40%7D%7D%2C%22plugins%22%3A%7B%22legend%22%3A%7B%22display%22%3Afalse%7D%2C%22title%22%3A%7B%22display%22%3Atrue%2C%22text%22%3A%22All%20504%22%2C%22font%22%3A%7B%22size%22%3A16%7D%7D%2C%22datalabels%22%3A%7B%22display%22%3Atrue%2C%22anchor%22%3A%22end%22%2C%22align%22%3A%22end%22%2C%22clamp%22%3Atrue%2C%22color%22%3A%22%23111827%22%2C%22font%22%3A%7B%22size%22%3A11%7D%2C%22formatter%22%3Afunction%28v%29%7Breturn%20v.toFixed%281%29%7D%7D%7D%2C%22scales%22%3A%7B%22x%22%3A%7B%22min%22%3A0%2C%22max%22%3A100%2C%22ticks%22%3A%7B%22stepSize%22%3A20%7D%2C%22title%22%3A%7B%22display%22%3Atrue%2C%22text%22%3A%22%25%22%7D%7D%2C%22y%22%3A%7B%22grid%22%3A%7B%22display%22%3Afalse%7D%7D%7D%7D%7D)

[![Easy](https://quickchart.io/chart?version=3&w=640&h=1210&bkg=white&c=%7B%22type%22%3A%22bar%22%2C%22data%22%3A%7B%22labels%22%3A%5B%22Jev%201.13.0%22%2C%22Lev%204B%22%2C%22Kev%204B%22%2C%22Tev1%204B%22%2C%22Tev1%200.8B%22%2C%22semif%22%2C%22decider-4b%20v2%22%2C%22JevK5%20v0.2%22%2C%22Plumb%204B%22%2C%22RSI-Jev%20v3.0%22%2C%22RSI-Jev%20v4.0-VL%22%2C%22Strands%202B%22%2C%22Clef-Flash%209B%22%2C%22AutoTrust%20JEV-9B%22%2C%22NeoHorse-Jev%204B%22%2C%22Mica%20v0.1%204B%22%2C%22Clef%2027B%22%2C%22AutoTrust%20JEV-27B%22%2C%22AutoJev%2027B%22%2C%22Rune%20v3%2026B-A4B%22%2C%22AutoTrust%20JEV-27B-VL%22%2C%22Jev%201.14.0%22%2C%22StandardOne%203B%22%2C%22d3-mini%204B%22%2C%22d3-flash%209B%22%2C%22d3%2027B%22%2C%22Shisa%20DE-2%22%2C%22AutoTrust%20v1.1%22%2C%22decider%2031B%22%2C%22pplx%20flash%204B%22%2C%22h2o%20lightning%204B%22%2C%22pplx%2027B%20v1.1%22%2C%22linne-oak%2031B%22%2C%22Jeff-1.0-Large%22%2C%22AutoTrust%20JEV-35B%22%2C%22Nimble%209B%20v3%22%2C%22StartLux%2027B%22%2C%22OpenAI%20gpt-6-luna%22%2C%22d3-nano%202B%22%2C%22d3-lite%200.8B%22%2C%22decider%200.8B%22%2C%22JevEmbed%204B%22%2C%22jeff%22%2C%22Laya%22%2C%22GLiNER%22%5D%2C%22datasets%22%3A%5B%7B%22data%22%3A%5B100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C97.2%2C97.2%2C93.1%2C83.3%2C62.5%2C52.8%2C33.3%5D%2C%22backgroundColor%22%3A%5B%22%230d9488%22%2C%22%237c3aed%22%2C%22%2392400e%22%2C%22%23ea580c%22%2C%22%230891b2%22%2C%22%234f46e5%22%2C%22%230f766e%22%2C%22%230369a1%22%2C%22%23b45309%22%2C%22%2338bdf8%22%2C%22%230ea5e9%22%2C%22%2384cc16%22%2C%22%2310b981%22%2C%22%23a21caf%22%2C%22%232563eb%22%2C%22%23dc2626%22%2C%22%23059669%22%2C%22%23c026d3%22%2C%22%23d97706%22%2C%22%23475569%22%2C%22%239333ea%22%2C%22%2314b8a6%22%2C%22%238b5cf6%22%2C%22%23f97316%22%2C%22%23ef4444%22%2C%22%23b91c1c%22%2C%22%2306b6d4%22%2C%22%23e879f9%22%2C%22%23115e59%22%2C%22%2364748b%22%2C%22%23facc15%22%2C%22%23334155%22%2C%22%234d7c0f%22%2C%22%23be185d%22%2C%22%237e22ce%22%2C%22%230e7490%22%2C%22%231d4ed8%22%2C%22%2310a37f%22%2C%22%23f59e0b%22%2C%22%23fdba74%22%2C%22%2316a34a%22%2C%22%23a16207%22%2C%22%23db2777%22%2C%22%23e11d48%22%2C%22%2365a30d%22%5D%2C%22borderWidth%22%3A0%2C%22barPercentage%22%3A0.82%7D%5D%7D%2C%22options%22%3A%7B%22indexAxis%22%3A%22y%22%2C%22layout%22%3A%7B%22padding%22%3A%7B%22right%22%3A40%7D%7D%2C%22plugins%22%3A%7B%22legend%22%3A%7B%22display%22%3Afalse%7D%2C%22title%22%3A%7B%22display%22%3Atrue%2C%22text%22%3A%22Easy%22%2C%22font%22%3A%7B%22size%22%3A16%7D%7D%2C%22datalabels%22%3A%7B%22display%22%3Atrue%2C%22anchor%22%3A%22end%22%2C%22align%22%3A%22end%22%2C%22clamp%22%3Atrue%2C%22color%22%3A%22%23111827%22%2C%22font%22%3A%7B%22size%22%3A11%7D%2C%22formatter%22%3Afunction%28v%29%7Breturn%20v.toFixed%281%29%7D%7D%7D%2C%22scales%22%3A%7B%22x%22%3A%7B%22min%22%3A0%2C%22max%22%3A100%2C%22ticks%22%3A%7B%22stepSize%22%3A20%7D%2C%22title%22%3A%7B%22display%22%3Atrue%2C%22text%22%3A%22%25%22%7D%7D%2C%22y%22%3A%7B%22grid%22%3A%7B%22display%22%3Afalse%7D%7D%7D%7D%7D)](https://quickchart.io/chart?version=3&w=640&h=1210&bkg=white&c=%7B%22type%22%3A%22bar%22%2C%22data%22%3A%7B%22labels%22%3A%5B%22Jev%201.13.0%22%2C%22Lev%204B%22%2C%22Kev%204B%22%2C%22Tev1%204B%22%2C%22Tev1%200.8B%22%2C%22semif%22%2C%22decider-4b%20v2%22%2C%22JevK5%20v0.2%22%2C%22Plumb%204B%22%2C%22RSI-Jev%20v3.0%22%2C%22RSI-Jev%20v4.0-VL%22%2C%22Strands%202B%22%2C%22Clef-Flash%209B%22%2C%22AutoTrust%20JEV-9B%22%2C%22NeoHorse-Jev%204B%22%2C%22Mica%20v0.1%204B%22%2C%22Clef%2027B%22%2C%22AutoTrust%20JEV-27B%22%2C%22AutoJev%2027B%22%2C%22Rune%20v3%2026B-A4B%22%2C%22AutoTrust%20JEV-27B-VL%22%2C%22Jev%201.14.0%22%2C%22StandardOne%203B%22%2C%22d3-mini%204B%22%2C%22d3-flash%209B%22%2C%22d3%2027B%22%2C%22Shisa%20DE-2%22%2C%22AutoTrust%20v1.1%22%2C%22decider%2031B%22%2C%22pplx%20flash%204B%22%2C%22h2o%20lightning%204B%22%2C%22pplx%2027B%20v1.1%22%2C%22linne-oak%2031B%22%2C%22Jeff-1.0-Large%22%2C%22AutoTrust%20JEV-35B%22%2C%22Nimble%209B%20v3%22%2C%22StartLux%2027B%22%2C%22OpenAI%20gpt-6-luna%22%2C%22d3-nano%202B%22%2C%22d3-lite%200.8B%22%2C%22decider%200.8B%22%2C%22JevEmbed%204B%22%2C%22jeff%22%2C%22Laya%22%2C%22GLiNER%22%5D%2C%22datasets%22%3A%5B%7B%22data%22%3A%5B100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C97.2%2C97.2%2C93.1%2C83.3%2C62.5%2C52.8%2C33.3%5D%2C%22backgroundColor%22%3A%5B%22%230d9488%22%2C%22%237c3aed%22%2C%22%2392400e%22%2C%22%23ea580c%22%2C%22%230891b2%22%2C%22%234f46e5%22%2C%22%230f766e%22%2C%22%230369a1%22%2C%22%23b45309%22%2C%22%2338bdf8%22%2C%22%230ea5e9%22%2C%22%2384cc16%22%2C%22%2310b981%22%2C%22%23a21caf%22%2C%22%232563eb%22%2C%22%23dc2626%22%2C%22%23059669%22%2C%22%23c026d3%22%2C%22%23d97706%22%2C%22%23475569%22%2C%22%239333ea%22%2C%22%2314b8a6%22%2C%22%238b5cf6%22%2C%22%23f97316%22%2C%22%23ef4444%22%2C%22%23b91c1c%22%2C%22%2306b6d4%22%2C%22%23e879f9%22%2C%22%23115e59%22%2C%22%2364748b%22%2C%22%23facc15%22%2C%22%23334155%22%2C%22%234d7c0f%22%2C%22%23be185d%22%2C%22%237e22ce%22%2C%22%230e7490%22%2C%22%231d4ed8%22%2C%22%2310a37f%22%2C%22%23f59e0b%22%2C%22%23fdba74%22%2C%22%2316a34a%22%2C%22%23a16207%22%2C%22%23db2777%22%2C%22%23e11d48%22%2C%22%2365a30d%22%5D%2C%22borderWidth%22%3A0%2C%22barPercentage%22%3A0.82%7D%5D%7D%2C%22options%22%3A%7B%22indexAxis%22%3A%22y%22%2C%22layout%22%3A%7B%22padding%22%3A%7B%22right%22%3A40%7D%7D%2C%22plugins%22%3A%7B%22legend%22%3A%7B%22display%22%3Afalse%7D%2C%22title%22%3A%7B%22display%22%3Atrue%2C%22text%22%3A%22Easy%22%2C%22font%22%3A%7B%22size%22%3A16%7D%7D%2C%22datalabels%22%3A%7B%22display%22%3Atrue%2C%22anchor%22%3A%22end%22%2C%22align%22%3A%22end%22%2C%22clamp%22%3Atrue%2C%22color%22%3A%22%23111827%22%2C%22font%22%3A%7B%22size%22%3A11%7D%2C%22formatter%22%3Afunction%28v%29%7Breturn%20v.toFixed%281%29%7D%7D%7D%2C%22scales%22%3A%7B%22x%22%3A%7B%22min%22%3A0%2C%22max%22%3A100%2C%22ticks%22%3A%7B%22stepSize%22%3A20%7D%2C%22title%22%3A%7B%22display%22%3Atrue%2C%22text%22%3A%22%25%22%7D%7D%2C%22y%22%3A%7B%22grid%22%3A%7B%22display%22%3Afalse%7D%7D%7D%7D%7D)

[![Medium](https://quickchart.io/chart?version=3&w=640&h=1210&bkg=white&c=%7B%22type%22%3A%22bar%22%2C%22data%22%3A%7B%22labels%22%3A%5B%22Jev%201.13.0%22%2C%22Tev1%204B%22%2C%22AutoTrust%20JEV-9B%22%2C%22Mica%20v0.1%204B%22%2C%22AutoTrust%20JEV-27B%22%2C%22AutoJev%2027B%22%2C%22AutoTrust%20JEV-27B-VL%22%2C%22Jev%201.14.0%22%2C%22d3%2027B%22%2C%22Shisa%20DE-2%22%2C%22AutoTrust%20v1.1%22%2C%22decider%2031B%22%2C%22pplx%2027B%20v1.1%22%2C%22linne-oak%2031B%22%2C%22Jeff-1.0-Large%22%2C%22Nimble%209B%20v3%22%2C%22StartLux%2027B%22%2C%22Kev%204B%22%2C%22decider-4b%20v2%22%2C%22Rune%20v3%2026B-A4B%22%2C%22AutoTrust%20JEV-35B%22%2C%22OpenAI%20gpt-6-luna%22%2C%22Clef%2027B%22%2C%22d3-mini%204B%22%2C%22d3-flash%209B%22%2C%22pplx%20flash%204B%22%2C%22h2o%20lightning%204B%22%2C%22NeoHorse-Jev%204B%22%2C%22Clef-Flash%209B%22%2C%22JevK5%20v0.2%22%2C%22Tev1%200.8B%22%2C%22RSI-Jev%20v4.0-VL%22%2C%22Plumb%204B%22%2C%22d3-nano%202B%22%2C%22semif%22%2C%22JevEmbed%204B%22%2C%22StandardOne%203B%22%2C%22RSI-Jev%20v3.0%22%2C%22d3-lite%200.8B%22%2C%22Lev%204B%22%2C%22Strands%202B%22%2C%22decider%200.8B%22%2C%22GLiNER%22%2C%22Laya%22%2C%22jeff%22%5D%2C%22datasets%22%3A%5B%7B%22data%22%3A%5B100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C99.0%2C99.0%2C99.0%2C99.0%2C99.0%2C97.9%2C97.9%2C97.9%2C96.9%2C96.9%2C95.8%2C93.8%2C91.7%2C89.6%2C87.5%2C86.5%2C86.5%2C85.4%2C85.4%2C85.4%2C80.2%2C80.2%2C79.2%2C76.0%2C72.9%2C40.6%2C38.5%2C35.4%5D%2C%22backgroundColor%22%3A%5B%22%230d9488%22%2C%22%23ea580c%22%2C%22%23a21caf%22%2C%22%23dc2626%22%2C%22%23c026d3%22%2C%22%23d97706%22%2C%22%239333ea%22%2C%22%2314b8a6%22%2C%22%23b91c1c%22%2C%22%2306b6d4%22%2C%22%23e879f9%22%2C%22%23115e59%22%2C%22%23334155%22%2C%22%234d7c0f%22%2C%22%23be185d%22%2C%22%230e7490%22%2C%22%231d4ed8%22%2C%22%2392400e%22%2C%22%230f766e%22%2C%22%23475569%22%2C%22%237e22ce%22%2C%22%2310a37f%22%2C%22%23059669%22%2C%22%23f97316%22%2C%22%23ef4444%22%2C%22%2364748b%22%2C%22%23facc15%22%2C%22%232563eb%22%2C%22%2310b981%22%2C%22%230369a1%22%2C%22%230891b2%22%2C%22%230ea5e9%22%2C%22%23b45309%22%2C%22%23f59e0b%22%2C%22%234f46e5%22%2C%22%23a16207%22%2C%22%238b5cf6%22%2C%22%2338bdf8%22%2C%22%23fdba74%22%2C%22%237c3aed%22%2C%22%2384cc16%22%2C%22%2316a34a%22%2C%22%2365a30d%22%2C%22%23e11d48%22%2C%22%23db2777%22%5D%2C%22borderWidth%22%3A0%2C%22barPercentage%22%3A0.82%7D%5D%7D%2C%22options%22%3A%7B%22indexAxis%22%3A%22y%22%2C%22layout%22%3A%7B%22padding%22%3A%7B%22right%22%3A40%7D%7D%2C%22plugins%22%3A%7B%22legend%22%3A%7B%22display%22%3Afalse%7D%2C%22title%22%3A%7B%22display%22%3Atrue%2C%22text%22%3A%22Medium%22%2C%22font%22%3A%7B%22size%22%3A16%7D%7D%2C%22datalabels%22%3A%7B%22display%22%3Atrue%2C%22anchor%22%3A%22end%22%2C%22align%22%3A%22end%22%2C%22clamp%22%3Atrue%2C%22color%22%3A%22%23111827%22%2C%22font%22%3A%7B%22size%22%3A11%7D%2C%22formatter%22%3Afunction%28v%29%7Breturn%20v.toFixed%281%29%7D%7D%7D%2C%22scales%22%3A%7B%22x%22%3A%7B%22min%22%3A0%2C%22max%22%3A100%2C%22ticks%22%3A%7B%22stepSize%22%3A20%7D%2C%22title%22%3A%7B%22display%22%3Atrue%2C%22text%22%3A%22%25%22%7D%7D%2C%22y%22%3A%7B%22grid%22%3A%7B%22display%22%3Afalse%7D%7D%7D%7D%7D)](https://quickchart.io/chart?version=3&w=640&h=1210&bkg=white&c=%7B%22type%22%3A%22bar%22%2C%22data%22%3A%7B%22labels%22%3A%5B%22Jev%201.13.0%22%2C%22Tev1%204B%22%2C%22AutoTrust%20JEV-9B%22%2C%22Mica%20v0.1%204B%22%2C%22AutoTrust%20JEV-27B%22%2C%22AutoJev%2027B%22%2C%22AutoTrust%20JEV-27B-VL%22%2C%22Jev%201.14.0%22%2C%22d3%2027B%22%2C%22Shisa%20DE-2%22%2C%22AutoTrust%20v1.1%22%2C%22decider%2031B%22%2C%22pplx%2027B%20v1.1%22%2C%22linne-oak%2031B%22%2C%22Jeff-1.0-Large%22%2C%22Nimble%209B%20v3%22%2C%22StartLux%2027B%22%2C%22Kev%204B%22%2C%22decider-4b%20v2%22%2C%22Rune%20v3%2026B-A4B%22%2C%22AutoTrust%20JEV-35B%22%2C%22OpenAI%20gpt-6-luna%22%2C%22Clef%2027B%22%2C%22d3-mini%204B%22%2C%22d3-flash%209B%22%2C%22pplx%20flash%204B%22%2C%22h2o%20lightning%204B%22%2C%22NeoHorse-Jev%204B%22%2C%22Clef-Flash%209B%22%2C%22JevK5%20v0.2%22%2C%22Tev1%200.8B%22%2C%22RSI-Jev%20v4.0-VL%22%2C%22Plumb%204B%22%2C%22d3-nano%202B%22%2C%22semif%22%2C%22JevEmbed%204B%22%2C%22StandardOne%203B%22%2C%22RSI-Jev%20v3.0%22%2C%22d3-lite%200.8B%22%2C%22Lev%204B%22%2C%22Strands%202B%22%2C%22decider%200.8B%22%2C%22GLiNER%22%2C%22Laya%22%2C%22jeff%22%5D%2C%22datasets%22%3A%5B%7B%22data%22%3A%5B100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C100.0%2C99.0%2C99.0%2C99.0%2C99.0%2C99.0%2C97.9%2C97.9%2C97.9%2C96.9%2C96.9%2C95.8%2C93.8%2C91.7%2C89.6%2C87.5%2C86.5%2C86.5%2C85.4%2C85.4%2C85.4%2C80.2%2C80.2%2C79.2%2C76.0%2C72.9%2C40.6%2C38.5%2C35.4%5D%2C%22backgroundColor%22%3A%5B%22%230d9488%22%2C%22%23ea580c%22%2C%22%23a21caf%22%2C%22%23dc2626%22%2C%22%23c026d3%22%2C%22%23d97706%22%2C%22%239333ea%22%2C%22%2314b8a6%22%2C%22%23b91c1c%22%2C%22%2306b6d4%22%2C%22%23e879f9%22%2C%22%23115e59%22%2C%22%23334155%22%2C%22%234d7c0f%22%2C%22%23be185d%22%2C%22%230e7490%22%2C%22%231d4ed8%22%2C%22%2392400e%22%2C%22%230f766e%22%2C%22%23475569%22%2C%22%237e22ce%22%2C%22%2310a37f%22%2C%22%23059669%22%2C%22%23f97316%22%2C%22%23ef4444%22%2C%22%2364748b%22%2C%22%23facc15%22%2C%22%232563eb%22%2C%22%2310b981%22%2C%22%230369a1%22%2C%22%230891b2%22%2C%22%230ea5e9%22%2C%22%23b45309%22%2C%22%23f59e0b%22%2C%22%234f46e5%22%2C%22%23a16207%22%2C%22%238b5cf6%22%2C%22%2338bdf8%22%2C%22%23fdba74%22%2C%22%237c3aed%22%2C%22%2384cc16%22%2C%22%2316a34a%22%2C%22%2365a30d%22%2C%22%23e11d48%22%2C%22%23db2777%22%5D%2C%22borderWidth%22%3A0%2C%22barPercentage%22%3A0.82%7D%5D%7D%2C%22options%22%3A%7B%22indexAxis%22%3A%22y%22%2C%22layout%22%3A%7B%22padding%22%3A%7B%22right%22%3A40%7D%7D%2C%22plugins%22%3A%7B%22legend%22%3A%7B%22display%22%3Afalse%7D%2C%22title%22%3A%7B%22display%22%3Atrue%2C%22text%22%3A%22Medium%22%2C%22font%22%3A%7B%22size%22%3A16%7D%7D%2C%22datalabels%22%3A%7B%22display%22%3Atrue%2C%22anchor%22%3A%22end%22%2C%22align%22%3A%22end%22%2C%22clamp%22%3Atrue%2C%22color%22%3A%22%23111827%22%2C%22font%22%3A%7B%22size%22%3A11%7D%2C%22formatter%22%3Afunction%28v%29%7Breturn%20v.toFixed%281%29%7D%7D%7D%2C%22scales%22%3A%7B%22x%22%3A%7B%22min%22%3A0%2C%22max%22%3A100%2C%22ticks%22%3A%7B%22stepSize%22%3A20%7D%2C%22title%22%3A%7B%22display%22%3Atrue%2C%22text%22%3A%22%25%22%7D%7D%2C%22y%22%3A%7B%22grid%22%3A%7B%22display%22%3Afalse%7D%7D%7D%7D%7D)

[![Hard](https://quickchart.io/chart?version=3&w=640&h=1210&bkg=white&c=%7B%22type%22%3A%22bar%22%2C%22data%22%3A%7B%22labels%22%3A%5B%22d3%2027B%22%2C%22AutoTrust%20v1.1%22%2C%22AutoTrust%20JEV-27B%22%2C%22AutoTrust%20JEV-27B-VL%22%2C%22pplx%2027B%20v1.1%22%2C%22d3-flash%209B%22%2C%22AutoTrust%20JEV-35B%22%2C%22Jev%201.13.0%22%2C%22linne-oak%2031B%22%2C%22StartLux%2027B%22%2C%22d3-mini%204B%22%2C%22Shisa%20DE-2%22%2C%22Jeff-1.0-Large%22%2C%22AutoJev%2027B%22%2C%22Rune%20v3%2026B-A4B%22%2C%22decider%2031B%22%2C%22d3-nano%202B%22%2C%22Jev%201.14.0%22%2C%22pplx%20flash%204B%22%2C%22Nimble%209B%20v3%22%2C%22OpenAI%20gpt-6-luna%22%2C%22h2o%20lightning%204B%22%2C%22decider-4b%20v2%22%2C%22NeoHorse-Jev%204B%22%2C%22Tev1%204B%22%2C%22AutoTrust%20JEV-9B%22%2C%22Clef%2027B%22%2C%22Mica%20v0.1%204B%22%2C%22Kev%204B%22%2C%22Clef-Flash%209B%22%2C%22RSI-Jev%20v4.0-VL%22%2C%22RSI-Jev%20v3.0%22%2C%22StandardOne%203B%22%2C%22Strands%202B%22%2C%22d3-lite%200.8B%22%2C%22GLiNER%22%2C%22JevEmbed%204B%22%2C%22semif%22%2C%22decider%200.8B%22%2C%22Tev1%200.8B%22%2C%22Lev%204B%22%2C%22Laya%22%2C%22jeff%22%2C%22JevK5%20v0.2%22%2C%22Plumb%204B%22%5D%2C%22datasets%22%3A%5B%7B%22data%22%3A%5B98.2%2C97.6%2C94.6%2C94.6%2C94.3%2C93.5%2C93.5%2C93.2%2C92.9%2C92.9%2C92.6%2C90.5%2C90.2%2C89.9%2C89.6%2C88.7%2C88.1%2C87.2%2C85.1%2C83.6%2C81.8%2C81.0%2C80.1%2C78.6%2C76.5%2C73.8%2C70.8%2C69.9%2C69.3%2C66.1%2C65.5%2C64.3%2C63.4%2C61.6%2C61.3%2C57.7%2C56.0%2C53.0%2C52.4%2C52.4%2C50.3%2C49.1%2C41.7%2C38.7%2C37.5%5D%2C%22backgroundColor%22%3A%5B%22%23b91c1c%22%2C%22%23e879f9%22%2C%22%23c026d3%22%2C%22%239333ea%22%2C%22%23334155%22%2C%22%23ef4444%22%2C%22%237e22ce%22%2C%22%230d9488%22%2C%22%234d7c0f%22%2C%22%231d4ed8%22%2C%22%23f97316%22%2C%22%2306b6d4%22%2C%22%23be185d%22%2C%22%23d97706%22%2C%22%23475569%22%2C%22%23115e59%22%2C%22%23f59e0b%22%2C%22%2314b8a6%22%2C%22%2364748b%22%2C%22%230e7490%22%2C%22%2310a37f%22%2C%22%23facc15%22%2C%22%230f766e%22%2C%22%232563eb%22%2C%22%23ea580c%22%2C%22%23a21caf%22%2C%22%23059669%22%2C%22%23dc2626%22%2C%22%2392400e%22%2C%22%2310b981%22%2C%22%230ea5e9%22%2C%22%2338bdf8%22%2C%22%238b5cf6%22%2C%22%2384cc16%22%2C%22%23fdba74%22%2C%22%2365a30d%22%2C%22%23a16207%22%2C%22%234f46e5%22%2C%22%2316a34a%22%2C%22%230891b2%22%2C%22%237c3aed%22%2C%22%23e11d48%22%2C%22%23db2777%22%2C%22%230369a1%22%2C%22%23b45309%22%5D%2C%22borderWidth%22%3A0%2C%22barPercentage%22%3A0.82%7D%5D%7D%2C%22options%22%3A%7B%22indexAxis%22%3A%22y%22%2C%22layout%22%3A%7B%22padding%22%3A%7B%22right%22%3A40%7D%7D%2C%22plugins%22%3A%7B%22legend%22%3A%7B%22display%22%3Afalse%7D%2C%22title%22%3A%7B%22display%22%3Atrue%2C%22text%22%3A%22Hard%22%2C%22font%22%3A%7B%22size%22%3A16%7D%7D%2C%22datalabels%22%3A%7B%22display%22%3Atrue%2C%22anchor%22%3A%22end%22%2C%22align%22%3A%22end%22%2C%22clamp%22%3Atrue%2C%22color%22%3A%22%23111827%22%2C%22font%22%3A%7B%22size%22%3A11%7D%2C%22formatter%22%3Afunction%28v%29%7Breturn%20v.toFixed%281%29%7D%7D%7D%2C%22scales%22%3A%7B%22x%22%3A%7B%22min%22%3A0%2C%22max%22%3A100%2C%22ticks%22%3A%7B%22stepSize%22%3A20%7D%2C%22title%22%3A%7B%22display%22%3Atrue%2C%22text%22%3A%22%25%22%7D%7D%2C%22y%22%3A%7B%22grid%22%3A%7B%22display%22%3Afalse%7D%7D%7D%7D%7D)](https://quickchart.io/chart?version=3&w=640&h=1210&bkg=white&c=%7B%22type%22%3A%22bar%22%2C%22data%22%3A%7B%22labels%22%3A%5B%22d3%2027B%22%2C%22AutoTrust%20v1.1%22%2C%22AutoTrust%20JEV-27B%22%2C%22AutoTrust%20JEV-27B-VL%22%2C%22pplx%2027B%20v1.1%22%2C%22d3-flash%209B%22%2C%22AutoTrust%20JEV-35B%22%2C%22Jev%201.13.0%22%2C%22linne-oak%2031B%22%2C%22StartLux%2027B%22%2C%22d3-mini%204B%22%2C%22Shisa%20DE-2%22%2C%22Jeff-1.0-Large%22%2C%22AutoJev%2027B%22%2C%22Rune%20v3%2026B-A4B%22%2C%22decider%2031B%22%2C%22d3-nano%202B%22%2C%22Jev%201.14.0%22%2C%22pplx%20flash%204B%22%2C%22Nimble%209B%20v3%22%2C%22OpenAI%20gpt-6-luna%22%2C%22h2o%20lightning%204B%22%2C%22decider-4b%20v2%22%2C%22NeoHorse-Jev%204B%22%2C%22Tev1%204B%22%2C%22AutoTrust%20JEV-9B%22%2C%22Clef%2027B%22%2C%22Mica%20v0.1%204B%22%2C%22Kev%204B%22%2C%22Clef-Flash%209B%22%2C%22RSI-Jev%20v4.0-VL%22%2C%22RSI-Jev%20v3.0%22%2C%22StandardOne%203B%22%2C%22Strands%202B%22%2C%22d3-lite%200.8B%22%2C%22GLiNER%22%2C%22JevEmbed%204B%22%2C%22semif%22%2C%22decider%200.8B%22%2C%22Tev1%200.8B%22%2C%22Lev%204B%22%2C%22Laya%22%2C%22jeff%22%2C%22JevK5%20v0.2%22%2C%22Plumb%204B%22%5D%2C%22datasets%22%3A%5B%7B%22data%22%3A%5B98.2%2C97.6%2C94.6%2C94.6%2C94.3%2C93.5%2C93.5%2C93.2%2C92.9%2C92.9%2C92.6%2C90.5%2C90.2%2C89.9%2C89.6%2C88.7%2C88.1%2C87.2%2C85.1%2C83.6%2C81.8%2C81.0%2C80.1%2C78.6%2C76.5%2C73.8%2C70.8%2C69.9%2C69.3%2C66.1%2C65.5%2C64.3%2C63.4%2C61.6%2C61.3%2C57.7%2C56.0%2C53.0%2C52.4%2C52.4%2C50.3%2C49.1%2C41.7%2C38.7%2C37.5%5D%2C%22backgroundColor%22%3A%5B%22%23b91c1c%22%2C%22%23e879f9%22%2C%22%23c026d3%22%2C%22%239333ea%22%2C%22%23334155%22%2C%22%23ef4444%22%2C%22%237e22ce%22%2C%22%230d9488%22%2C%22%234d7c0f%22%2C%22%231d4ed8%22%2C%22%23f97316%22%2C%22%2306b6d4%22%2C%22%23be185d%22%2C%22%23d97706%22%2C%22%23475569%22%2C%22%23115e59%22%2C%22%23f59e0b%22%2C%22%2314b8a6%22%2C%22%2364748b%22%2C%22%230e7490%22%2C%22%2310a37f%22%2C%22%23facc15%22%2C%22%230f766e%22%2C%22%232563eb%22%2C%22%23ea580c%22%2C%22%23a21caf%22%2C%22%23059669%22%2C%22%23dc2626%22%2C%22%2392400e%22%2C%22%2310b981%22%2C%22%230ea5e9%22%2C%22%2338bdf8%22%2C%22%238b5cf6%22%2C%22%2384cc16%22%2C%22%23fdba74%22%2C%22%2365a30d%22%2C%22%23a16207%22%2C%22%234f46e5%22%2C%22%2316a34a%22%2C%22%230891b2%22%2C%22%237c3aed%22%2C%22%23e11d48%22%2C%22%23db2777%22%2C%22%230369a1%22%2C%22%23b45309%22%5D%2C%22borderWidth%22%3A0%2C%22barPercentage%22%3A0.82%7D%5D%7D%2C%22options%22%3A%7B%22indexAxis%22%3A%22y%22%2C%22layout%22%3A%7B%22padding%22%3A%7B%22right%22%3A40%7D%7D%2C%22plugins%22%3A%7B%22legend%22%3A%7B%22display%22%3Afalse%7D%2C%22title%22%3A%7B%22display%22%3Atrue%2C%22text%22%3A%22Hard%22%2C%22font%22%3A%7B%22size%22%3A16%7D%7D%2C%22datalabels%22%3A%7B%22display%22%3Atrue%2C%22anchor%22%3A%22end%22%2C%22align%22%3A%22end%22%2C%22clamp%22%3Atrue%2C%22color%22%3A%22%23111827%22%2C%22font%22%3A%7B%22size%22%3A11%7D%2C%22formatter%22%3Afunction%28v%29%7Breturn%20v.toFixed%281%29%7D%7D%7D%2C%22scales%22%3A%7B%22x%22%3A%7B%22min%22%3A0%2C%22max%22%3A100%2C%22ticks%22%3A%7B%22stepSize%22%3A20%7D%2C%22title%22%3A%7B%22display%22%3Atrue%2C%22text%22%3A%22%25%22%7D%7D%2C%22y%22%3A%7B%22grid%22%3A%7B%22display%22%3Afalse%7D%7D%7D%7D%7D)

| Model | Easy /72 | Medium /96 | Hard /336 | All /504 | Hard priority reversals /29 |
| --- | ---: | ---: | ---: | ---: | ---: |
| d3 27B | 72 | 96 | 330 | 498 | 29 |
| AutoTrust JEV-27B v1.1 | 72 | 96 | 328 | 496 | 29 |
| AutoTrust JEV-27B | 72 | 96 | 318 | 486 | 28 |
| AutoTrust JEV-27B-VL | 72 | 96 | 318 | 486 | 28 |
| pplx-decider v1.1 27B | 72 | 96 | 317 | 485 | 28 |
| Jev `jev-1.13.0` | 72 | 96 | 313 | 481 | 28 |
| AutoTrust JEV-35B | 72 | 95 | 314 | 481 | 27 |
| d3-flash 9B | 72 | 94 | 314 | 480 | 28 |
| linne-oak 1.0 31B | 72 | 96 | 312 | 480 | 28 |
| StartLux v1.1 27B Instant | 72 | 96 | 312 | 480 | 29 |
| d3-mini 4B | 72 | 94 | 311 | 477 | 28 |
| Shisa DE-2 26B-A4B | 72 | 96 | 304 | 472 | 26 |
| Jeff-1.0-Large 31B | 72 | 96 | 303 | 471 | 29 |
| AutoJev-27B | 72 | 96 | 302 | 470 | 29 |
| Rune v3 26B-A4B | 72 | 95 | 301 | 468 | 24 |
| decider-31b | 72 | 96 | 298 | 466 | 24 |
| Jev `jev-1.14.0` | 72 | 96 | 293 | 461 | 24 |
| pplx-decider-flash v1.2 4B | 72 | 93 | 286 | 451 | 21 |
| d3-nano 2B | 70 | 83 | 296 | 449 | 25 |
| Bespoke Nimble 9B v3 | 72 | 96 | 281 | 449 | 29 |
| OpenAI `gpt-6-luna` (hosted) | 72 | 95 | 275 | 442 | 23 |
| h2o-lightning-4b | 72 | 93 | 272 | 437 | 14 |
| decider-4b v2 | 72 | 95 | 269 | 436 | 14 |
| NeoHorse-Jev-4B | 72 | 92 | 264 | 428 | 5 |
| Tev1-4B | 72 | 96 | 257 | 425 | 15 |
| AutoTrust JEV-9B | 72 | 96 | 248 | 416 | 4 |
| Clef 27B | 72 | 94 | 238 | 404 | 4 |
| Mica v0.1 4B | 72 | 96 | 235 | 403 | 7 |
| Kev 4B | 72 | 95 | 233 | 400 | 9 |
| Clef-Flash 9B | 72 | 90 | 222 | 384 | 3 |
| RSI-Jev v4.0-VL 2B | 72 | 84 | 220 | 376 | 0 |
| StandardOne-3B-SH v2 | 72 | 82 | 213 | 367 | 1 |
| RSI-Jev v3.0 2B | 72 | 77 | 216 | 365 | 2 |
| d3-lite 0.8B | 70 | 77 | 206 | 353 | 6 |
| Strands Decider 2B | 72 | 73 | 207 | 352 | 1 |
| Tev1-0.8B | 72 | 86 | 176 | 334 | 0 |
| SemIf, Qwen3.5-4B | 72 | 82 | 178 | 332 | 2 |
| JevEmbed Qwen3-Embedding-4B | 60 | 82 | 188 | 330 | 1 |
| Lev 4B | 72 | 76 | 169 | 317 | 1 |
| decider-0.8b | 67 | 70 | 176 | 313 | 1 |
| JevK5 4B v0.2 | 72 | 88 | 130 | 290 | 2 |
| Plumb-4B | 72 | 83 | 126 | 281 | 1 |
| GLiNER2.5-Decide | 24 | 39 | 194 | 257 | 0 |
| Laya multilingual | 38 | 37 | 165 | 240 | 0 |
| jeff (GLiFormer large) | 45 | 34 | 140 | 219 | 0 |

Easy and medium are regression and "insufficient information" controls.
Hard carries the difference: a priority reversal counts only if both ends
of a same-case pair are right in both orders. Always answering B already
scores 194/336 on hard. All 22,176 first attempts (44 models × 504) are
valid, with no errors. Adapters differ, so these are task-level scores, not
weight comparisons. Jev `jev-1.14.0` scores 20 fewer than `jev-1.13.0`, all
on hard (293 vs 313; a repeat gave 462 in all). From 0.0.7 jev-lite pins
`jev-1.14.0` anyway (owner decision): pin `jev-1.13.0` yourself while it is
served if hard text decisions matter most. Self-host pick:
[d3-nano](https://huggingface.co/vllm-sr/d3-nano) (2.2B, Apache-2.0, 449/504),
which `LocalTypeSafe` runs (above). This bank ships in the `jev-lite`
0.0.2–0.0.4 sdists, so models released after 2026-10-03 may have seen it.

Earlier banks (`semantic-decisions@v0.0.1`, the controlled slices, r2) are
retired. v0.0.1 survives only at package tag `v0.0.1`, and none of their
scores is pooled with v0.0.2. The project redistributes no measured model's
weights, except the Apache-2.0 d3-nano export that `LocalTypeSafe`
downloads, which carries its LICENSE and NOTICE.

## Latency by question count

One request, one state, N questions (1 → 255), on one machine (NVIDIA GB10,
local models on its GPU, `LocalTypeSafe` on its CPU; hosted rows over the
network). Speed depends on the hardware, so read the growth, not the
milliseconds. The state is 24 joined bank policies (about 2,000 tokens);
p50 of three uncached requests (each opens with a fresh request number, so
no server reuses an earlier prefix). A backend that scores one question per call
gets them one after another, as its client would. "cut" means a smaller N
already passed 45 s or ran out of memory (local models get at most 70% of the
GPU memory). Questions are numbered, so no two in a request are the same.
The charts show one model for each way of reading the state; the table lists
every model.

*State read* is inferred from the growth: a model that reads the state once
per question takes about N times as long for N questions. Hosted rows also
ran 384 and 512 questions (Jev 1.14: 458 and 528 ms); Jev answered 768
questions (about 37k tokens) with HTTP 400.

[![Latency by question count](https://quickchart.io/chart?version=3&w=900&h=640&bkg=white&c=%7B%22type%22%3A%22line%22%2C%22data%22%3A%7B%22datasets%22%3A%5B%7B%22label%22%3A%22Jev%201.14.0%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A174.1%7D%2C%7B%22x%22%3A2%2C%22y%22%3A173.5%7D%2C%7B%22x%22%3A4%2C%22y%22%3A174.4%7D%2C%7B%22x%22%3A8%2C%22y%22%3A176.9%7D%2C%7B%22x%22%3A16%2C%22y%22%3A178.7%7D%2C%7B%22x%22%3A32%2C%22y%22%3A195.8%7D%2C%7B%22x%22%3A64%2C%22y%22%3A213.5%7D%2C%7B%22x%22%3A128%2C%22y%22%3A276.9%7D%2C%7B%22x%22%3A255%2C%22y%22%3A335.3%7D%2C%7B%22x%22%3A384%2C%22y%22%3A457.9%7D%2C%7B%22x%22%3A512%2C%22y%22%3A527.8%7D%5D%2C%22borderColor%22%3A%22%2314b8a6%22%2C%22backgroundColor%22%3A%22%2314b8a6%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22OpenAI%20gpt-6-luna%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A222.2%7D%2C%7B%22x%22%3A2%2C%22y%22%3A244.5%7D%2C%7B%22x%22%3A4%2C%22y%22%3A255.2%7D%2C%7B%22x%22%3A8%2C%22y%22%3A319.0%7D%2C%7B%22x%22%3A16%2C%22y%22%3A306.2%7D%2C%7B%22x%22%3A32%2C%22y%22%3A373.0%7D%2C%7B%22x%22%3A64%2C%22y%22%3A511.3%7D%2C%7B%22x%22%3A128%2C%22y%22%3A645.0%7D%2C%7B%22x%22%3A255%2C%22y%22%3A770.4%7D%2C%7B%22x%22%3A384%2C%22y%22%3A957.7%7D%5D%2C%22borderColor%22%3A%22%2310a37f%22%2C%22backgroundColor%22%3A%22%2310a37f%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22LocalTypeSafe%200.0.6%20CPU%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A4573.1%7D%2C%7B%22x%22%3A2%2C%22y%22%3A9059.3%7D%2C%7B%22x%22%3A4%2C%22y%22%3A18423.7%7D%2C%7B%22x%22%3A8%2C%22y%22%3A36608.4%7D%2C%7B%22x%22%3A16%2C%22y%22%3A72931.9%7D%5D%2C%22borderColor%22%3A%22%2364748b%22%2C%22backgroundColor%22%3A%22%2364748b%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22LocalTypeSafe%200.0.7%20CPU%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A4627.4%7D%2C%7B%22x%22%3A2%2C%22y%22%3A4858.7%7D%2C%7B%22x%22%3A4%2C%22y%22%3A5315.1%7D%2C%7B%22x%22%3A8%2C%22y%22%3A5967.3%7D%2C%7B%22x%22%3A16%2C%22y%22%3A7484.0%7D%2C%7B%22x%22%3A32%2C%22y%22%3A10423.8%7D%2C%7B%22x%22%3A64%2C%22y%22%3A16184.3%7D%2C%7B%22x%22%3A128%2C%22y%22%3A28285.7%7D%2C%7B%22x%22%3A255%2C%22y%22%3A52292.1%7D%5D%2C%22borderColor%22%3A%22%231e293b%22%2C%22backgroundColor%22%3A%22%231e293b%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22d3%2027B%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A2139.8%7D%2C%7B%22x%22%3A2%2C%22y%22%3A4171.4%7D%2C%7B%22x%22%3A4%2C%22y%22%3A8418.9%7D%2C%7B%22x%22%3A8%2C%22y%22%3A23784.6%7D%2C%7B%22x%22%3A16%2C%22y%22%3A33332.6%7D%2C%7B%22x%22%3A32%2C%22y%22%3A66171.7%7D%5D%2C%22borderColor%22%3A%22%23b91c1c%22%2C%22backgroundColor%22%3A%22%23b91c1c%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22d3%2027B%20state%20once%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A2359.7%7D%2C%7B%22x%22%3A2%2C%22y%22%3A2409.2%7D%2C%7B%22x%22%3A4%2C%22y%22%3A2530.0%7D%2C%7B%22x%22%3A8%2C%22y%22%3A2777.9%7D%2C%7B%22x%22%3A16%2C%22y%22%3A3489.2%7D%2C%7B%22x%22%3A32%2C%22y%22%3A5118.7%7D%2C%7B%22x%22%3A64%2C%22y%22%3A8096.6%7D%2C%7B%22x%22%3A128%2C%22y%22%3A15165.0%7D%2C%7B%22x%22%3A255%2C%22y%22%3A26821.8%7D%5D%2C%22borderColor%22%3A%22%237f1d1d%22%2C%22backgroundColor%22%3A%22%237f1d1d%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22Tev1%204B%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A620.1%7D%2C%7B%22x%22%3A2%2C%22y%22%3A1238.8%7D%2C%7B%22x%22%3A4%2C%22y%22%3A2481.1%7D%2C%7B%22x%22%3A8%2C%22y%22%3A5041.0%7D%2C%7B%22x%22%3A16%2C%22y%22%3A9956.0%7D%2C%7B%22x%22%3A32%2C%22y%22%3A19988.5%7D%2C%7B%22x%22%3A64%2C%22y%22%3A40006.7%7D%2C%7B%22x%22%3A128%2C%22y%22%3A80532.1%7D%5D%2C%22borderColor%22%3A%22%23ea580c%22%2C%22backgroundColor%22%3A%22%23ea580c%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22Kev%204B%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A115.0%7D%2C%7B%22x%22%3A2%2C%22y%22%3A123.4%7D%2C%7B%22x%22%3A4%2C%22y%22%3A137.2%7D%2C%7B%22x%22%3A8%2C%22y%22%3A183.8%7D%2C%7B%22x%22%3A16%2C%22y%22%3A292.6%7D%2C%7B%22x%22%3A32%2C%22y%22%3A505.3%7D%2C%7B%22x%22%3A64%2C%22y%22%3A975.6%7D%2C%7B%22x%22%3A128%2C%22y%22%3A2019.2%7D%2C%7B%22x%22%3A255%2C%22y%22%3A4316.0%7D%5D%2C%22borderColor%22%3A%22%2392400e%22%2C%22backgroundColor%22%3A%22%2392400e%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22StandardOne%203B%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A424.7%7D%2C%7B%22x%22%3A2%2C%22y%22%3A435.9%7D%2C%7B%22x%22%3A4%2C%22y%22%3A451.9%7D%2C%7B%22x%22%3A8%2C%22y%22%3A506.7%7D%2C%7B%22x%22%3A16%2C%22y%22%3A595.7%7D%2C%7B%22x%22%3A32%2C%22y%22%3A818.6%7D%2C%7B%22x%22%3A64%2C%22y%22%3A1288.2%7D%2C%7B%22x%22%3A128%2C%22y%22%3A2363.5%7D%2C%7B%22x%22%3A255%2C%22y%22%3A5057.3%7D%5D%2C%22borderColor%22%3A%22%238b5cf6%22%2C%22backgroundColor%22%3A%22%238b5cf6%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22Nimble%209B%20v3%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A1078.5%7D%2C%7B%22x%22%3A2%2C%22y%22%3A1226.6%7D%2C%7B%22x%22%3A4%2C%22y%22%3A1322.5%7D%2C%7B%22x%22%3A8%2C%22y%22%3A1582.3%7D%2C%7B%22x%22%3A16%2C%22y%22%3A2101.7%7D%2C%7B%22x%22%3A32%2C%22y%22%3A3398.3%7D%2C%7B%22x%22%3A64%2C%22y%22%3A6468.2%7D%2C%7B%22x%22%3A128%2C%22y%22%3A15269.8%7D%2C%7B%22x%22%3A255%2C%22y%22%3A47205.9%7D%5D%2C%22borderColor%22%3A%22%230e7490%22%2C%22backgroundColor%22%3A%22%230e7490%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%5D%7D%2C%22options%22%3A%7B%22plugins%22%3A%7B%22title%22%3A%7B%22display%22%3Atrue%2C%22text%22%3A%22Latency%20by%20question%20count%22%2C%22font%22%3A%7B%22size%22%3A16%7D%7D%2C%22legend%22%3A%7B%22position%22%3A%22right%22%2C%22labels%22%3A%7B%22boxWidth%22%3A10%2C%22font%22%3A%7B%22size%22%3A9%7D%7D%7D%7D%2C%22scales%22%3A%7B%22x%22%3A%7B%22type%22%3A%22logarithmic%22%2C%22title%22%3A%7B%22display%22%3Atrue%2C%22text%22%3A%22questions%20in%20one%20request%22%7D%7D%2C%22y%22%3A%7B%22type%22%3A%22logarithmic%22%2C%22title%22%3A%7B%22display%22%3Atrue%2C%22text%22%3A%22ms%20per%20request%20%28p50%29%22%7D%7D%7D%7D%7D)](https://quickchart.io/chart?version=3&w=900&h=640&bkg=white&c=%7B%22type%22%3A%22line%22%2C%22data%22%3A%7B%22datasets%22%3A%5B%7B%22label%22%3A%22Jev%201.14.0%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A174.1%7D%2C%7B%22x%22%3A2%2C%22y%22%3A173.5%7D%2C%7B%22x%22%3A4%2C%22y%22%3A174.4%7D%2C%7B%22x%22%3A8%2C%22y%22%3A176.9%7D%2C%7B%22x%22%3A16%2C%22y%22%3A178.7%7D%2C%7B%22x%22%3A32%2C%22y%22%3A195.8%7D%2C%7B%22x%22%3A64%2C%22y%22%3A213.5%7D%2C%7B%22x%22%3A128%2C%22y%22%3A276.9%7D%2C%7B%22x%22%3A255%2C%22y%22%3A335.3%7D%2C%7B%22x%22%3A384%2C%22y%22%3A457.9%7D%2C%7B%22x%22%3A512%2C%22y%22%3A527.8%7D%5D%2C%22borderColor%22%3A%22%2314b8a6%22%2C%22backgroundColor%22%3A%22%2314b8a6%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22OpenAI%20gpt-6-luna%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A222.2%7D%2C%7B%22x%22%3A2%2C%22y%22%3A244.5%7D%2C%7B%22x%22%3A4%2C%22y%22%3A255.2%7D%2C%7B%22x%22%3A8%2C%22y%22%3A319.0%7D%2C%7B%22x%22%3A16%2C%22y%22%3A306.2%7D%2C%7B%22x%22%3A32%2C%22y%22%3A373.0%7D%2C%7B%22x%22%3A64%2C%22y%22%3A511.3%7D%2C%7B%22x%22%3A128%2C%22y%22%3A645.0%7D%2C%7B%22x%22%3A255%2C%22y%22%3A770.4%7D%2C%7B%22x%22%3A384%2C%22y%22%3A957.7%7D%5D%2C%22borderColor%22%3A%22%2310a37f%22%2C%22backgroundColor%22%3A%22%2310a37f%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22LocalTypeSafe%200.0.6%20CPU%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A4573.1%7D%2C%7B%22x%22%3A2%2C%22y%22%3A9059.3%7D%2C%7B%22x%22%3A4%2C%22y%22%3A18423.7%7D%2C%7B%22x%22%3A8%2C%22y%22%3A36608.4%7D%2C%7B%22x%22%3A16%2C%22y%22%3A72931.9%7D%5D%2C%22borderColor%22%3A%22%2364748b%22%2C%22backgroundColor%22%3A%22%2364748b%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22LocalTypeSafe%200.0.7%20CPU%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A4627.4%7D%2C%7B%22x%22%3A2%2C%22y%22%3A4858.7%7D%2C%7B%22x%22%3A4%2C%22y%22%3A5315.1%7D%2C%7B%22x%22%3A8%2C%22y%22%3A5967.3%7D%2C%7B%22x%22%3A16%2C%22y%22%3A7484.0%7D%2C%7B%22x%22%3A32%2C%22y%22%3A10423.8%7D%2C%7B%22x%22%3A64%2C%22y%22%3A16184.3%7D%2C%7B%22x%22%3A128%2C%22y%22%3A28285.7%7D%2C%7B%22x%22%3A255%2C%22y%22%3A52292.1%7D%5D%2C%22borderColor%22%3A%22%231e293b%22%2C%22backgroundColor%22%3A%22%231e293b%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22d3%2027B%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A2139.8%7D%2C%7B%22x%22%3A2%2C%22y%22%3A4171.4%7D%2C%7B%22x%22%3A4%2C%22y%22%3A8418.9%7D%2C%7B%22x%22%3A8%2C%22y%22%3A23784.6%7D%2C%7B%22x%22%3A16%2C%22y%22%3A33332.6%7D%2C%7B%22x%22%3A32%2C%22y%22%3A66171.7%7D%5D%2C%22borderColor%22%3A%22%23b91c1c%22%2C%22backgroundColor%22%3A%22%23b91c1c%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22d3%2027B%20state%20once%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A2359.7%7D%2C%7B%22x%22%3A2%2C%22y%22%3A2409.2%7D%2C%7B%22x%22%3A4%2C%22y%22%3A2530.0%7D%2C%7B%22x%22%3A8%2C%22y%22%3A2777.9%7D%2C%7B%22x%22%3A16%2C%22y%22%3A3489.2%7D%2C%7B%22x%22%3A32%2C%22y%22%3A5118.7%7D%2C%7B%22x%22%3A64%2C%22y%22%3A8096.6%7D%2C%7B%22x%22%3A128%2C%22y%22%3A15165.0%7D%2C%7B%22x%22%3A255%2C%22y%22%3A26821.8%7D%5D%2C%22borderColor%22%3A%22%237f1d1d%22%2C%22backgroundColor%22%3A%22%237f1d1d%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22Tev1%204B%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A620.1%7D%2C%7B%22x%22%3A2%2C%22y%22%3A1238.8%7D%2C%7B%22x%22%3A4%2C%22y%22%3A2481.1%7D%2C%7B%22x%22%3A8%2C%22y%22%3A5041.0%7D%2C%7B%22x%22%3A16%2C%22y%22%3A9956.0%7D%2C%7B%22x%22%3A32%2C%22y%22%3A19988.5%7D%2C%7B%22x%22%3A64%2C%22y%22%3A40006.7%7D%2C%7B%22x%22%3A128%2C%22y%22%3A80532.1%7D%5D%2C%22borderColor%22%3A%22%23ea580c%22%2C%22backgroundColor%22%3A%22%23ea580c%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22Kev%204B%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A115.0%7D%2C%7B%22x%22%3A2%2C%22y%22%3A123.4%7D%2C%7B%22x%22%3A4%2C%22y%22%3A137.2%7D%2C%7B%22x%22%3A8%2C%22y%22%3A183.8%7D%2C%7B%22x%22%3A16%2C%22y%22%3A292.6%7D%2C%7B%22x%22%3A32%2C%22y%22%3A505.3%7D%2C%7B%22x%22%3A64%2C%22y%22%3A975.6%7D%2C%7B%22x%22%3A128%2C%22y%22%3A2019.2%7D%2C%7B%22x%22%3A255%2C%22y%22%3A4316.0%7D%5D%2C%22borderColor%22%3A%22%2392400e%22%2C%22backgroundColor%22%3A%22%2392400e%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22StandardOne%203B%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A424.7%7D%2C%7B%22x%22%3A2%2C%22y%22%3A435.9%7D%2C%7B%22x%22%3A4%2C%22y%22%3A451.9%7D%2C%7B%22x%22%3A8%2C%22y%22%3A506.7%7D%2C%7B%22x%22%3A16%2C%22y%22%3A595.7%7D%2C%7B%22x%22%3A32%2C%22y%22%3A818.6%7D%2C%7B%22x%22%3A64%2C%22y%22%3A1288.2%7D%2C%7B%22x%22%3A128%2C%22y%22%3A2363.5%7D%2C%7B%22x%22%3A255%2C%22y%22%3A5057.3%7D%5D%2C%22borderColor%22%3A%22%238b5cf6%22%2C%22backgroundColor%22%3A%22%238b5cf6%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22Nimble%209B%20v3%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A1078.5%7D%2C%7B%22x%22%3A2%2C%22y%22%3A1226.6%7D%2C%7B%22x%22%3A4%2C%22y%22%3A1322.5%7D%2C%7B%22x%22%3A8%2C%22y%22%3A1582.3%7D%2C%7B%22x%22%3A16%2C%22y%22%3A2101.7%7D%2C%7B%22x%22%3A32%2C%22y%22%3A3398.3%7D%2C%7B%22x%22%3A64%2C%22y%22%3A6468.2%7D%2C%7B%22x%22%3A128%2C%22y%22%3A15269.8%7D%2C%7B%22x%22%3A255%2C%22y%22%3A47205.9%7D%5D%2C%22borderColor%22%3A%22%230e7490%22%2C%22backgroundColor%22%3A%22%230e7490%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%5D%7D%2C%22options%22%3A%7B%22plugins%22%3A%7B%22title%22%3A%7B%22display%22%3Atrue%2C%22text%22%3A%22Latency%20by%20question%20count%22%2C%22font%22%3A%7B%22size%22%3A16%7D%7D%2C%22legend%22%3A%7B%22position%22%3A%22right%22%2C%22labels%22%3A%7B%22boxWidth%22%3A10%2C%22font%22%3A%7B%22size%22%3A9%7D%7D%7D%7D%2C%22scales%22%3A%7B%22x%22%3A%7B%22type%22%3A%22logarithmic%22%2C%22title%22%3A%7B%22display%22%3Atrue%2C%22text%22%3A%22questions%20in%20one%20request%22%7D%7D%2C%22y%22%3A%7B%22type%22%3A%22logarithmic%22%2C%22title%22%3A%7B%22display%22%3Atrue%2C%22text%22%3A%22ms%20per%20request%20%28p50%29%22%7D%7D%7D%7D%7D)

[![Latency growth from one question](https://quickchart.io/chart?version=3&w=900&h=640&bkg=white&c=%7B%22type%22%3A%22line%22%2C%22data%22%3A%7B%22datasets%22%3A%5B%7B%22label%22%3A%22Jev%201.14.0%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A1.0%7D%2C%7B%22x%22%3A2%2C%22y%22%3A1.0%7D%2C%7B%22x%22%3A4%2C%22y%22%3A1.0%7D%2C%7B%22x%22%3A8%2C%22y%22%3A1.02%7D%2C%7B%22x%22%3A16%2C%22y%22%3A1.03%7D%2C%7B%22x%22%3A32%2C%22y%22%3A1.12%7D%2C%7B%22x%22%3A64%2C%22y%22%3A1.23%7D%2C%7B%22x%22%3A128%2C%22y%22%3A1.59%7D%2C%7B%22x%22%3A255%2C%22y%22%3A1.93%7D%2C%7B%22x%22%3A384%2C%22y%22%3A2.63%7D%2C%7B%22x%22%3A512%2C%22y%22%3A3.03%7D%5D%2C%22borderColor%22%3A%22%2314b8a6%22%2C%22backgroundColor%22%3A%22%2314b8a6%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22OpenAI%20gpt-6-luna%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A1.0%7D%2C%7B%22x%22%3A2%2C%22y%22%3A1.1%7D%2C%7B%22x%22%3A4%2C%22y%22%3A1.15%7D%2C%7B%22x%22%3A8%2C%22y%22%3A1.44%7D%2C%7B%22x%22%3A16%2C%22y%22%3A1.38%7D%2C%7B%22x%22%3A32%2C%22y%22%3A1.68%7D%2C%7B%22x%22%3A64%2C%22y%22%3A2.3%7D%2C%7B%22x%22%3A128%2C%22y%22%3A2.9%7D%2C%7B%22x%22%3A255%2C%22y%22%3A3.47%7D%2C%7B%22x%22%3A384%2C%22y%22%3A4.31%7D%5D%2C%22borderColor%22%3A%22%2310a37f%22%2C%22backgroundColor%22%3A%22%2310a37f%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22LocalTypeSafe%200.0.6%20CPU%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A1.0%7D%2C%7B%22x%22%3A2%2C%22y%22%3A1.98%7D%2C%7B%22x%22%3A4%2C%22y%22%3A4.03%7D%2C%7B%22x%22%3A8%2C%22y%22%3A8.01%7D%2C%7B%22x%22%3A16%2C%22y%22%3A15.95%7D%5D%2C%22borderColor%22%3A%22%2364748b%22%2C%22backgroundColor%22%3A%22%2364748b%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22LocalTypeSafe%200.0.7%20CPU%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A1.0%7D%2C%7B%22x%22%3A2%2C%22y%22%3A1.05%7D%2C%7B%22x%22%3A4%2C%22y%22%3A1.15%7D%2C%7B%22x%22%3A8%2C%22y%22%3A1.29%7D%2C%7B%22x%22%3A16%2C%22y%22%3A1.62%7D%2C%7B%22x%22%3A32%2C%22y%22%3A2.25%7D%2C%7B%22x%22%3A64%2C%22y%22%3A3.5%7D%2C%7B%22x%22%3A128%2C%22y%22%3A6.11%7D%2C%7B%22x%22%3A255%2C%22y%22%3A11.3%7D%5D%2C%22borderColor%22%3A%22%231e293b%22%2C%22backgroundColor%22%3A%22%231e293b%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22d3%2027B%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A1.0%7D%2C%7B%22x%22%3A2%2C%22y%22%3A1.95%7D%2C%7B%22x%22%3A4%2C%22y%22%3A3.93%7D%2C%7B%22x%22%3A8%2C%22y%22%3A11.12%7D%2C%7B%22x%22%3A16%2C%22y%22%3A15.58%7D%2C%7B%22x%22%3A32%2C%22y%22%3A30.92%7D%5D%2C%22borderColor%22%3A%22%23b91c1c%22%2C%22backgroundColor%22%3A%22%23b91c1c%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22d3%2027B%20state%20once%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A1.0%7D%2C%7B%22x%22%3A2%2C%22y%22%3A1.02%7D%2C%7B%22x%22%3A4%2C%22y%22%3A1.07%7D%2C%7B%22x%22%3A8%2C%22y%22%3A1.18%7D%2C%7B%22x%22%3A16%2C%22y%22%3A1.48%7D%2C%7B%22x%22%3A32%2C%22y%22%3A2.17%7D%2C%7B%22x%22%3A64%2C%22y%22%3A3.43%7D%2C%7B%22x%22%3A128%2C%22y%22%3A6.43%7D%2C%7B%22x%22%3A255%2C%22y%22%3A11.37%7D%5D%2C%22borderColor%22%3A%22%237f1d1d%22%2C%22backgroundColor%22%3A%22%237f1d1d%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22Tev1%204B%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A1.0%7D%2C%7B%22x%22%3A2%2C%22y%22%3A2.0%7D%2C%7B%22x%22%3A4%2C%22y%22%3A4.0%7D%2C%7B%22x%22%3A8%2C%22y%22%3A8.13%7D%2C%7B%22x%22%3A16%2C%22y%22%3A16.06%7D%2C%7B%22x%22%3A32%2C%22y%22%3A32.23%7D%2C%7B%22x%22%3A64%2C%22y%22%3A64.52%7D%2C%7B%22x%22%3A128%2C%22y%22%3A129.87%7D%5D%2C%22borderColor%22%3A%22%23ea580c%22%2C%22backgroundColor%22%3A%22%23ea580c%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22Kev%204B%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A1.0%7D%2C%7B%22x%22%3A2%2C%22y%22%3A1.07%7D%2C%7B%22x%22%3A4%2C%22y%22%3A1.19%7D%2C%7B%22x%22%3A8%2C%22y%22%3A1.6%7D%2C%7B%22x%22%3A16%2C%22y%22%3A2.54%7D%2C%7B%22x%22%3A32%2C%22y%22%3A4.39%7D%2C%7B%22x%22%3A64%2C%22y%22%3A8.48%7D%2C%7B%22x%22%3A128%2C%22y%22%3A17.56%7D%2C%7B%22x%22%3A255%2C%22y%22%3A37.53%7D%5D%2C%22borderColor%22%3A%22%2392400e%22%2C%22backgroundColor%22%3A%22%2392400e%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22StandardOne%203B%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A1.0%7D%2C%7B%22x%22%3A2%2C%22y%22%3A1.03%7D%2C%7B%22x%22%3A4%2C%22y%22%3A1.06%7D%2C%7B%22x%22%3A8%2C%22y%22%3A1.19%7D%2C%7B%22x%22%3A16%2C%22y%22%3A1.4%7D%2C%7B%22x%22%3A32%2C%22y%22%3A1.93%7D%2C%7B%22x%22%3A64%2C%22y%22%3A3.03%7D%2C%7B%22x%22%3A128%2C%22y%22%3A5.57%7D%2C%7B%22x%22%3A255%2C%22y%22%3A11.91%7D%5D%2C%22borderColor%22%3A%22%238b5cf6%22%2C%22backgroundColor%22%3A%22%238b5cf6%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22Nimble%209B%20v3%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A1.0%7D%2C%7B%22x%22%3A2%2C%22y%22%3A1.14%7D%2C%7B%22x%22%3A4%2C%22y%22%3A1.23%7D%2C%7B%22x%22%3A8%2C%22y%22%3A1.47%7D%2C%7B%22x%22%3A16%2C%22y%22%3A1.95%7D%2C%7B%22x%22%3A32%2C%22y%22%3A3.15%7D%2C%7B%22x%22%3A64%2C%22y%22%3A6.0%7D%2C%7B%22x%22%3A128%2C%22y%22%3A14.16%7D%2C%7B%22x%22%3A255%2C%22y%22%3A43.77%7D%5D%2C%22borderColor%22%3A%22%230e7490%22%2C%22backgroundColor%22%3A%22%230e7490%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%5D%7D%2C%22options%22%3A%7B%22plugins%22%3A%7B%22title%22%3A%7B%22display%22%3Atrue%2C%22text%22%3A%22Latency%20growth%20from%20one%20question%22%2C%22font%22%3A%7B%22size%22%3A16%7D%7D%2C%22legend%22%3A%7B%22position%22%3A%22right%22%2C%22labels%22%3A%7B%22boxWidth%22%3A10%2C%22font%22%3A%7B%22size%22%3A9%7D%7D%7D%7D%2C%22scales%22%3A%7B%22x%22%3A%7B%22type%22%3A%22logarithmic%22%2C%22title%22%3A%7B%22display%22%3Atrue%2C%22text%22%3A%22questions%20in%20one%20request%22%7D%7D%2C%22y%22%3A%7B%22type%22%3A%22logarithmic%22%2C%22title%22%3A%7B%22display%22%3Atrue%2C%22text%22%3A%22times%20the%201-question%20latency%22%7D%7D%7D%7D%7D)](https://quickchart.io/chart?version=3&w=900&h=640&bkg=white&c=%7B%22type%22%3A%22line%22%2C%22data%22%3A%7B%22datasets%22%3A%5B%7B%22label%22%3A%22Jev%201.14.0%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A1.0%7D%2C%7B%22x%22%3A2%2C%22y%22%3A1.0%7D%2C%7B%22x%22%3A4%2C%22y%22%3A1.0%7D%2C%7B%22x%22%3A8%2C%22y%22%3A1.02%7D%2C%7B%22x%22%3A16%2C%22y%22%3A1.03%7D%2C%7B%22x%22%3A32%2C%22y%22%3A1.12%7D%2C%7B%22x%22%3A64%2C%22y%22%3A1.23%7D%2C%7B%22x%22%3A128%2C%22y%22%3A1.59%7D%2C%7B%22x%22%3A255%2C%22y%22%3A1.93%7D%2C%7B%22x%22%3A384%2C%22y%22%3A2.63%7D%2C%7B%22x%22%3A512%2C%22y%22%3A3.03%7D%5D%2C%22borderColor%22%3A%22%2314b8a6%22%2C%22backgroundColor%22%3A%22%2314b8a6%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22OpenAI%20gpt-6-luna%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A1.0%7D%2C%7B%22x%22%3A2%2C%22y%22%3A1.1%7D%2C%7B%22x%22%3A4%2C%22y%22%3A1.15%7D%2C%7B%22x%22%3A8%2C%22y%22%3A1.44%7D%2C%7B%22x%22%3A16%2C%22y%22%3A1.38%7D%2C%7B%22x%22%3A32%2C%22y%22%3A1.68%7D%2C%7B%22x%22%3A64%2C%22y%22%3A2.3%7D%2C%7B%22x%22%3A128%2C%22y%22%3A2.9%7D%2C%7B%22x%22%3A255%2C%22y%22%3A3.47%7D%2C%7B%22x%22%3A384%2C%22y%22%3A4.31%7D%5D%2C%22borderColor%22%3A%22%2310a37f%22%2C%22backgroundColor%22%3A%22%2310a37f%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22LocalTypeSafe%200.0.6%20CPU%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A1.0%7D%2C%7B%22x%22%3A2%2C%22y%22%3A1.98%7D%2C%7B%22x%22%3A4%2C%22y%22%3A4.03%7D%2C%7B%22x%22%3A8%2C%22y%22%3A8.01%7D%2C%7B%22x%22%3A16%2C%22y%22%3A15.95%7D%5D%2C%22borderColor%22%3A%22%2364748b%22%2C%22backgroundColor%22%3A%22%2364748b%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22LocalTypeSafe%200.0.7%20CPU%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A1.0%7D%2C%7B%22x%22%3A2%2C%22y%22%3A1.05%7D%2C%7B%22x%22%3A4%2C%22y%22%3A1.15%7D%2C%7B%22x%22%3A8%2C%22y%22%3A1.29%7D%2C%7B%22x%22%3A16%2C%22y%22%3A1.62%7D%2C%7B%22x%22%3A32%2C%22y%22%3A2.25%7D%2C%7B%22x%22%3A64%2C%22y%22%3A3.5%7D%2C%7B%22x%22%3A128%2C%22y%22%3A6.11%7D%2C%7B%22x%22%3A255%2C%22y%22%3A11.3%7D%5D%2C%22borderColor%22%3A%22%231e293b%22%2C%22backgroundColor%22%3A%22%231e293b%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22d3%2027B%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A1.0%7D%2C%7B%22x%22%3A2%2C%22y%22%3A1.95%7D%2C%7B%22x%22%3A4%2C%22y%22%3A3.93%7D%2C%7B%22x%22%3A8%2C%22y%22%3A11.12%7D%2C%7B%22x%22%3A16%2C%22y%22%3A15.58%7D%2C%7B%22x%22%3A32%2C%22y%22%3A30.92%7D%5D%2C%22borderColor%22%3A%22%23b91c1c%22%2C%22backgroundColor%22%3A%22%23b91c1c%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22d3%2027B%20state%20once%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A1.0%7D%2C%7B%22x%22%3A2%2C%22y%22%3A1.02%7D%2C%7B%22x%22%3A4%2C%22y%22%3A1.07%7D%2C%7B%22x%22%3A8%2C%22y%22%3A1.18%7D%2C%7B%22x%22%3A16%2C%22y%22%3A1.48%7D%2C%7B%22x%22%3A32%2C%22y%22%3A2.17%7D%2C%7B%22x%22%3A64%2C%22y%22%3A3.43%7D%2C%7B%22x%22%3A128%2C%22y%22%3A6.43%7D%2C%7B%22x%22%3A255%2C%22y%22%3A11.37%7D%5D%2C%22borderColor%22%3A%22%237f1d1d%22%2C%22backgroundColor%22%3A%22%237f1d1d%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22Tev1%204B%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A1.0%7D%2C%7B%22x%22%3A2%2C%22y%22%3A2.0%7D%2C%7B%22x%22%3A4%2C%22y%22%3A4.0%7D%2C%7B%22x%22%3A8%2C%22y%22%3A8.13%7D%2C%7B%22x%22%3A16%2C%22y%22%3A16.06%7D%2C%7B%22x%22%3A32%2C%22y%22%3A32.23%7D%2C%7B%22x%22%3A64%2C%22y%22%3A64.52%7D%2C%7B%22x%22%3A128%2C%22y%22%3A129.87%7D%5D%2C%22borderColor%22%3A%22%23ea580c%22%2C%22backgroundColor%22%3A%22%23ea580c%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22Kev%204B%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A1.0%7D%2C%7B%22x%22%3A2%2C%22y%22%3A1.07%7D%2C%7B%22x%22%3A4%2C%22y%22%3A1.19%7D%2C%7B%22x%22%3A8%2C%22y%22%3A1.6%7D%2C%7B%22x%22%3A16%2C%22y%22%3A2.54%7D%2C%7B%22x%22%3A32%2C%22y%22%3A4.39%7D%2C%7B%22x%22%3A64%2C%22y%22%3A8.48%7D%2C%7B%22x%22%3A128%2C%22y%22%3A17.56%7D%2C%7B%22x%22%3A255%2C%22y%22%3A37.53%7D%5D%2C%22borderColor%22%3A%22%2392400e%22%2C%22backgroundColor%22%3A%22%2392400e%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22StandardOne%203B%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A1.0%7D%2C%7B%22x%22%3A2%2C%22y%22%3A1.03%7D%2C%7B%22x%22%3A4%2C%22y%22%3A1.06%7D%2C%7B%22x%22%3A8%2C%22y%22%3A1.19%7D%2C%7B%22x%22%3A16%2C%22y%22%3A1.4%7D%2C%7B%22x%22%3A32%2C%22y%22%3A1.93%7D%2C%7B%22x%22%3A64%2C%22y%22%3A3.03%7D%2C%7B%22x%22%3A128%2C%22y%22%3A5.57%7D%2C%7B%22x%22%3A255%2C%22y%22%3A11.91%7D%5D%2C%22borderColor%22%3A%22%238b5cf6%22%2C%22backgroundColor%22%3A%22%238b5cf6%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%2C%7B%22label%22%3A%22Nimble%209B%20v3%22%2C%22data%22%3A%5B%7B%22x%22%3A1%2C%22y%22%3A1.0%7D%2C%7B%22x%22%3A2%2C%22y%22%3A1.14%7D%2C%7B%22x%22%3A4%2C%22y%22%3A1.23%7D%2C%7B%22x%22%3A8%2C%22y%22%3A1.47%7D%2C%7B%22x%22%3A16%2C%22y%22%3A1.95%7D%2C%7B%22x%22%3A32%2C%22y%22%3A3.15%7D%2C%7B%22x%22%3A64%2C%22y%22%3A6.0%7D%2C%7B%22x%22%3A128%2C%22y%22%3A14.16%7D%2C%7B%22x%22%3A255%2C%22y%22%3A43.77%7D%5D%2C%22borderColor%22%3A%22%230e7490%22%2C%22backgroundColor%22%3A%22%230e7490%22%2C%22fill%22%3Afalse%2C%22pointRadius%22%3A2%2C%22borderWidth%22%3A1.5%7D%5D%7D%2C%22options%22%3A%7B%22plugins%22%3A%7B%22title%22%3A%7B%22display%22%3Atrue%2C%22text%22%3A%22Latency%20growth%20from%20one%20question%22%2C%22font%22%3A%7B%22size%22%3A16%7D%7D%2C%22legend%22%3A%7B%22position%22%3A%22right%22%2C%22labels%22%3A%7B%22boxWidth%22%3A10%2C%22font%22%3A%7B%22size%22%3A9%7D%7D%7D%7D%2C%22scales%22%3A%7B%22x%22%3A%7B%22type%22%3A%22logarithmic%22%2C%22title%22%3A%7B%22display%22%3Atrue%2C%22text%22%3A%22questions%20in%20one%20request%22%7D%7D%2C%22y%22%3A%7B%22type%22%3A%22logarithmic%22%2C%22title%22%3A%7B%22display%22%3Atrue%2C%22text%22%3A%22times%20the%201-question%20latency%22%7D%7D%7D%7D%7D)

| Model | State read | 1 | 16 | 64 | 255 | 255 vs 1 | Largest N |
| --- | --- | ---: | ---: | ---: | ---: | ---: | ---: |
| Jev `jev-1.13.0` | hosted | 185 | 190 | 220 | 327 | 1.8x | 512 |
| Jev `jev-1.14.0` | hosted | 174 | 179 | 214 | 335 | 1.9x | 512 |
| OpenAI `gpt-6-luna` (hosted) | hosted | 222 | 306 | 511 | 770 | 3.5x | 384 |
| `LocalTypeSafe` 0.0.7 (CPU) | once | 4,627 | 7,484 | 16,184 | 52,292 | 11.3x | 255 |
| d3 27B, state once | once | 2,360 | 3,489 | 8,097 | 26,822 | 11.4x | 255 |
| d3-flash 9B, state once | once | 744 | 1,132 | 2,712 | 8,616 | 11.6x | 255 |
| StandardOne-3B-SH v2 | once | 425 | 596 | 1,288 | 5,057 | 11.9x | 255 |
| d3-mini 4B, state once | once | 506 | 788 | 1,918 | 6,221 | 12.3x | 255 |
| d3-nano 2B, state once | once | 216 | 327 | 836 | 2,720 | 12.6x | 255 |
| RSI-Jev v3.0 2B | once | 187 | 297 | 653 | cut | — | 64 |
| RSI-Jev v4.0-VL 2B | once | 187 | 296 | 651 | cut | — | 64 |
| Clef 27B | once | 2,154 | 3,606 | 8,824 | cut | — | 128 |
| d3-lite 0.8B, state once | once | 140 | 227 | 602 | 2,009 | 14.4x | 255 |
| decider-31b | once | 1,346 | 3,068 | 6,095 | 19,722 | 14.6x | 255 |
| Clef-Flash 9B | once | 710 | 1,224 | 2,898 | cut | — | 128 |
| Strands Decider 2B | once | 357 | 596 | 1,792 | 7,184 | 20.1x | 255 |
| Lev 4B, prefix fork (not its default) | once | 814 | 1,822 | 5,212 | cut | — | 128 |
| pplx-decider-flash v1.2 4B | once | 464 | 699 | 1,481 | 12,304 | 26.5x | 255 |
| Kev 4B | once, cut to 384 tokens | 115 | 293 | 976 | 4,316 | 37.5x | 255 |
| Bespoke Nimble 9B v3 | once | 1,078 | 2,102 | 6,468 | 47,206 | 43.8x | 255 |
| JevEmbed Qwen3-Embedding-4B | once | 850 | 4,306 | 15,660 | 56,498 | 66.5x | 255 |
| AutoTrust JEV-35B | per question | 737 | 6,300 | 24,226 | cut | — | 128 |
| h2o-lightning-4b | per question | 274 | 3,921 | 15,466 | 60,958 | 222.7x | 255 |
| AutoTrust JEV-27B v1.1 | per question | 1,519 | 22,456 | 88,019 | cut | — | 64 |
| linne-oak 1.0 31B | per question | 3,285 | cut | cut | cut | — | 8 |
| d3 27B | per question | 2,140 | 33,333 | cut | cut | — | 32 |
| d3-flash 9B | per question | 664 | 10,503 | 41,828 | cut | — | 128 |
| Tev1-0.8B | per question | 163 | 2,596 | 10,370 | 41,297 | 252.9x | 255 |
| `LocalTypeSafe` 0.0.5–0.0.6 (CPU) | per question | 4,573 | 72,932 | cut | cut | — | 16 |
| Jeff-1.0-Large 31B | per question | 3,317 | 53,070 | cut | cut | — | 16 |
| decider-4b v2 | per question | 468 | 7,358 | 29,721 | cut | — | 128 |
| SemIf, Qwen3.5-4B | per question | 454 | 7,254 | 29,053 | cut | — | 128 |
| pplx-decider v1.1 27B | per question | 2,139 | 34,239 | cut | cut | — | 32 |
| Mica v0.1 4B | per question | 463 | 7,408 | 29,669 | cut | — | 128 |
| AutoTrust JEV-27B | per question | 2,077 | 33,371 | cut | cut | — | 32 |
| Rune v3 26B-A4B | per question | 1,169 | 18,788 | 75,194 | cut | — | 64 |
| AutoTrust JEV-9B | per question | 665 | 10,708 | 42,942 | cut | — | 128 |
| AutoTrust JEV-27B-VL | per question | 2,042 | 32,852 | cut | cut | — | 32 |
| JevK5 4B v0.2 | per question | 488 | 7,840 | 31,604 | cut | — | 128 |
| Shisa DE-2 26B-A4B | per question | 2,405 | 38,762 | cut | cut | — | 32 |
| jeff (GLiFormer large) | per question | 2,405 | 38,556 | cut | cut | — | 32 |
| AutoJev-27B | per question | 2,071 | 33,331 | cut | cut | — | 32 |
| GLiNER2.5-Decide | per question | 9,867 | cut | cut | cut | — | 8 |
| Plumb-4B | per question | 483 | 7,776 | 31,105 | cut | — | 128 |
| d3-mini 4B | per question | 458 | 7,398 | 29,616 | cut | — | 128 |
| Tev1-4B | per question | 620 | 9,956 | 40,007 | cut | — | 128 |
| NeoHorse-Jev-4B | per question | 432 | 7,169 | cut | cut | — | 16 |
| Laya multilingual | per question | 20 | 346 | 1,350 | 5,287 | 271.1x | 255 |
| d3-nano 2B | per question | 194 | 3,334 | 13,379 | 53,317 | 274.1x | 255 |
| StartLux v1.1 27B Instant | per question | 1,489 | 28,838 | cut | cut | — | 32 |
| d3-lite 0.8B | per question | 124 | 2,287 | 9,133 | 36,344 | 293.6x | 255 |
| decider-0.8b | per question | 126 | 2,375 | 9,345 | 37,224 | 296.4x | 255 |
| Lev 4B | per question | 1,505 | 24,501 | cut | cut | — | 32 |

"State once" rows serve the d3 checkpoints with the state read once and the
questions batched on it; in d3's noncausal layers the state then no longer
sees the questions. The score on `semantic-decisions@v0.0.2` (/504) changes
only for some sizes: nano 449 → 449, lite 353 → 342, mini 477 → 459, 27B
498 → 496 (flash: 67 → 67 on a 72-slot sample). jev-lite 0.0.7 ships only
the nano one, as `LocalTypeSafe`.

## License

Apache-2.0
