# Is VINCO SOTA?

Pending benchmarks. Announced Nov 5, 2026; no independent results yet.

As of Nov 22, 2026, VINCO is pending benchmarks on isitsota: announced Nov 5, 2026, with no independent results on the boards we track yet.

Distributed Cognition's specialist model for competitive 1v1 Catan. Trained with offline reinforcement learning on simulated games and self-play fine-tuning, without human-game data, at about 0.5M parameters. Public playable demo at catan.dcog.com (X sign-in to play); technical report at dcog.com/research/solving-1v1-catan. No general API or open weights stated.

Canonical: https://isitsota.com/m/vinco
Last updated: 2026-11-05
Tracking since: 2026-11-05

## Announced facts

| Fact | Value |
| --- | --- |
| Lab | Distributed Cognition |
| Announced | Nov 5, 2026 |
| API | Not yet · None stated |
| Open weights | Closed · None stated |
| Parameters | ~0.5M |
| Role | Specialist 1v1 Catan agent (rush to 15 victory points, 30-second turns, no trading between players) |
| Training | Offline RL on simulated games plus self-play fine-tuning; lab says no training on human games; under a day of training on two RTX PRO 6000 Blackwell GPUs |
| Access | Public playable demo at https://catan.dcog.com (X sign-in to play); VINCO rating fixed at 3000 on the demo leaderboard |
| Report | https://dcog.com/research/solving-1v1-catan/ |

Tags marked unsure are not stated on the lab's own pages.

## What would let us call it

Verdicts use only independent boards we already track. Two consecutive snapshots are required before any crown.

- [x] 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)

## What the lab claims

Self-reported. Shown for context; not used in any verdict.

- Lab claim: on November 1, 2026, two distributed accounts held 1st and 2nd place globally on Colonist.io's 1v1 Catan leaderboard with under 24 hours of total gameplay (secret deployment; public intro Nov 5).
- Lab claim: across its games, VINCO wins when opponents are behind by a third of the game points, with an average lead of 5 victory points.
- Lab claim: ~0.5M-parameter VINCO scored 36/40 wins in the lab's matches; Opus 5.5 1/10, Fable 5.1 1/10, Astra 2/10, Kimi K3 0/10 (each opponent 10 games vs VINCO; high reasoning).

## Sources

- LAB: [solving 1v1 catan (Distributed Cognition)](https://dcog.com/research/solving-1v1-catan/) (2026-11-05)
- LAB: [Play Catan against VINCO (demo)](https://catan.dcog.com) (2026-11-05)
- LAB: [Distributed Cognition on X: VINCO](https://x.com/distcog/status/2118402967808487546) (2026-11-05)

Tracking ends Feb 3, 2027 unless the API or weights become available.

## Related

- Lab: [Distributed Cognition](https://isitsota.com/labs/distributed-cognition.md)
- All pending releases: https://isitsota.com/releases.md
- JSON: https://api.isitsota.com/v1/models/vinco
