Inkling debuts at #9 among open-weight models in Agent Arena
arena · x · 2026-07-21
Inkling by Thinking Machines debuts at #9 among open-weight models and #30 overall in Agent Arena.
- It is described as the best open model from the US for long-running agentic tasks, and the only US model in the top 10 open-weight field.
- Weaknesses: user sentiment is poor (-18.0%, #38), steerability is weak (#35), and task completion is only #29.
- Bright spot: it does relatively well at recovering from CLI errors (+6.4%, #16).
- The post notes that if Kimi K3 weights ship on schedule by July 27, it could take the top open-weight slot.
Related event: Thinking Machines' Inkling Debuts at 9th in Open-Weight on Agent Arena(3 posts)→
More from Models
- Kimi K2.8 Preview rolls out: near-K3 coding performance, 1M context for all tiers — teortaxesTex · 2026-09-11
- Looking for a classifier of software engineering task shapes to pick models per task — StewartalsopIII · 2026-09-11
- DeepSeek V4 Pro API to continue after Sept 2026, billing unchanged — teortaxesTex · 2026-09-11
- DeepSeek V4.1 Flash Hits 98% of GPT-6 Astra's Score at 1.4% of the Cost in Third-Party Benchmark — ayushtweetshere · 2026-09-11
- TheZvi Polls: Has Your Coding Model Choice Changed Since Fable 5.1 and Astra? — TheZvi · 2026-09-11
- antirez Weighs In on Anthropic Banning Minors From Using Claude — antirez · 2026-09-11