linne-oak-1.0 is a 31B decision model. Give it a state (text, a document, JSON, a conversation or an agent trajectory)
and a question with a set of options, and it returns one option together with a probability for every option, in a single
forward pass.
It is the larger sibling of linne-fern-1.0: linne-fern is a 27B causal model
that also chats, linne-oak is a dedicated decision model.
Highlights
- Decision Index public score 63.94, measured with the official kit (edition 0.3, all 37 public benchmarks, 140,620 requests,
complete).
- Reads the whole prompt at once: the full-attention layers attend in both directions over the prompt (sliding-window layers
keep their native behavior), so the decision sees the state, the question and every option together.
- Strong across the board: knowledge, language understanding, retrieval and classification, tools and automation.
- Agent-safe: a strong step-level guard for agents with a low false-block rate.
- Fast: one forward pass per decision; median 63 ms per request on one H200.
linne-oak-1.0 is a decision-only model. Because its full-attention layers are bidirectional, it does not generate text:
use it to pick options and read probabilities, not as a chat model. For chat, use the base model or linne-fern-1.0.
Decision Index (public part, edition 0.3)
Chance-corrected skill, higher is better. linne rows measured by us with the official kit; the other rows are the public-part
scores published on the leaderboard.
| Model |
Overall |
Knowledge |
Language |
Retrieval |
Tools |
Arts |
| linne-oak-1.0 (31B) |
63.94 |
54.3 |
66.6 |
69.1 |
78.5 |
45.1 |
| linne-fern-1.0 (27B) |
58.92 |
44.9 |
62.8 |
62.4 |
79.0 |
41.5 |
| Perplexity Decider v1.1 (27B), leaderboard |
62.25 |
| Quyet-1.0-Large (31B), leaderboard |
60.79 |
Only the public part (20% of the full leaderboard score) can be measured outside the leaderboard.
Usage
The prompt is one user message (chat template, thinking off): the state (compact JSON when not a string), a blank line,
Question: <instructions>, one line <label>: <name> - <description> per option, and
Answer with the letter of one option only. The answer is the softmax over the next-token logits of the option letters
(with the model's final logit soft-capping), read at the last prompt position, with the full-attention layers bidirectional.
linne_engine.py (with nc_hook.py) in this repository implements exactly this, as a Decision Index engine and as a library:
import sys
from huggingface_hub import snapshot_download
path = snapshot_download("MISA-core/linne-oak-1-0")
sys.path.insert(0, path)
from linne_engine import LinneEngine
model = LinneEngine(model=path, device="cuda:0", noncausal=True)
result, _ = model(
"My Stripe integration keeps failing. Please help ASAP.",
{"team": {"type": "choice", "instructions": "Which team should handle this request?",
"criteria": {"billing": "Charges and refunds", "technical_support": "Integration errors",
"sales": "Questions about buying a product"}}},
)
print(result["answers"]["team"]) # chosen option and a probability for every option
Run the Decision Index public suite:
PYTHONPATH=<path> python -m decision_index pipeline --engine linne_engine:LinneEngine \
--option model=<path> --option device=cuda:0 --option noncausal=1 --out runs/linne-oak-1-0
BF16 weights are about 62 GB and fit one 80 GB GPU.
License
Apache 2.0, following the base model's license.
