Pending benchmarksTracking since Oct 18, 2026
Is OrcaCyber Nano 1.0 state of the art?Pending benchmarks.Announced Oct 18, 2026; no independent results yet.
A 27B open-weight vision-capable model built for defensive cybersecurity work such as vulnerability research and incident response.
Announced facts
The details so far
- Lab
- OrcaRouter
- Announced
- Oct 18, 2026
- API
- Not yet
- Open weights
- Published
- Parameters
- 27B
- Focus
- Defensive security: reverse engineering, vulnerability analysis, secure code review, triage, incident response
- Context
- 262K tokens
- Input
- Text and images (vision)
- Runs
- Fully offline on your own GPU; GGUF builds for 16 GB and 24 GB cards
- Weights
- Hugging Face, gated with automatic approval
- License
- Other (custom)
- Series
- OrcaCyber: Nano, Zero and Ultra. OrcaCyber Zero is API-only on OrcaRouter.
Tags marked unsure are not stated on the lab's own pages.
Next steps
What would let us call it
Verdicts use only independent boards we already track. Two consecutive snapshots are required before any crown.
- Publicly announced with a source
- Usable by the public (API or weights)
- Listed on an independent board we track
- Enough results for a rank in a category
- Holds across two snapshots (crowns only)
Self-reported
What the lab claims
Self-reported. Shown for context; not used in any verdict.
Lab claim: the most capable local model in its under-16GB class, based on early evals.
Lab-reported scores (self-reported, harness still being finalized): AIME 2025 96.23%, MATH-500 98.60%, GSM8K 97.50%, MBPP+ 84.66%, IFEval 82.81%.
Receipts
Sources
LABOrcaRouter announcement on Xx.com/OrcaRouter/status/2111871993532580058 · Oct 18, 2026REPOOrcaCyber-Nano-1.0 on Hugging Facehuggingface.co/orcarouter/OrcaCyber-Nano-1.0 · Oct 18, 2026Tracking continues until an independent board lists it.