0

Codenames

Frontier

PrimeIntellect Codenames environment for evaluation and RL training

Open
Domain
rl-env
License
unknown
Published
Mar 2026
Canonical
app.primeintellect.ai/dashboard/environments/maxbittker/codenames

Cite

Notes

Only stored in your browser.

Attribution

Leaderboard scores
prime-hub
Attribution policy →

Top score 2.08 by Claude Opus 4.6 - 9 models reporting (6 frontier)

Score history

8
00.751.52.253Apr 25Jul 25Oct 25Jan 26GPT-4.1GPT-5 MiniClaude Opus 4.6

Top models

9
CodenamesBar chart with 9 bars. Highest value: Claude Opus 4.6 at 2.1.
9 models

Related tools

1
View all

Implementations, trainers, datasets and scaffolds linked to this eval.

Codenames RL Env (Community)

Official

PrimeIntellect Codenames environment for evaluation and RL training

ImplementationRL EnvGames

FAQ

What is Codenames?
PrimeIntellect Codenames environment for evaluation and RL training
What is the current top score on Codenames?
The top reported score is 2.08 by Claude Opus 4.6, across 9 models reporting (6 from frontier labs).
How can a model improve its Codenames score?
Tools linked to Codenames on Sophon include Codenames RL Env (Community) - RL environments, datasets, and scaffolds that target this eval.
What license is Codenames under?
Codenames is available under unknown.