Pending benchmarksTracking since Oct 16, 2026
Is Bonsai 2 state of the art?Pending benchmarks.Announced Oct 15, 2026; no independent results yet.
PrismML's ternary on-device models in 2B, 4B and 9B sizes, based on Qwen3.5, with reasoning, vision and tool calling. Footprints are 0.43GB, 0.95GB and 2.0GB. Open weights under Apache 2.0; the lab mentions access through the Darkbloom API.
Announced facts
The details so far
- Lab
- PrismML
- Announced
- Oct 15, 2026
- API
- Live
- Open weights
- Published
- Parameters
- 2B, 4B, 9B
- Base model
- Qwen3.5
- Precision
- Ternary
- Footprint
- 0.43GB (2B), 0.95GB (4B), 2.0GB (9B)
- License
- Apache 2.0
- API
- Darkbloom APIunsure · mentioned in the lab thread
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)
Receipts
Sources
LABPrismML announcement on Xx.com/PrismML/status/2110823964650164348 · Oct 15, 2026Tracking continues until an independent board lists it.