Thinky Machines Inkling Debuts at No. 9 on Agent Arena Open-Weight Leaderboard
Thinky Machines’ Inkling AI model debuted at No. 9 on the Agent Arena leaderboard for open-weight models, ranking 30th overall. The result marks the highest placement for a US-developed open model in the category.
The model’s aggregate performance score increased 9.6%, driven by a 16th-place finish in recovering from command-line interface errors. It ranked lower on user sentiment metrics, placing 38th in praise versus complaint at a -18.0% spread and 35th in steerability. The leaderboard evaluates models across millions of real-world, long-horizon agentic tasks that utilize web search, filesystem, and terminal tools.
From the sources (3 posts)
@arenaInkling by @thinkymachines debuts at #9 in Agent Arena among open-weight models, and #30 overall. This makes the best open model from the US for agentic long running tasks and also the only US model in the top 10 open model field. If Kimi K
@arenaInkling is the #9 open model, ranking #30 overall in the Agent Arena (+9.6%) - #18 Bash Recovery (+6.4%) - #29 Confirmed Task Success (-5.6%) - #35 Tool Hallucination (-0.9%) - #35 Steerability (-10.4%) - #38 Praise vs. Complaint (-18.0%)
@ml_angelopoulosRT @arena: Inkling by @thinkymachines debuts at #9 in Agent Arena among open-weight models, and #30 overall. This makes the best open model…