Thinking Machines releases Inkling 2 Small
What happened
On November 6, Inkling 2 Small appeared on the Hacker News front page. The model uses a mixture-of-experts architecture with 594B total parameters and 17B active parameters. It accepts image input and lets users adjust how much reasoning effort it spends. Training used large-scale unified reinforcement learning across coding, vision, knowledge work and reasoning tasks. Inkling 2 Small scored 31 on the Artificial Analysis Intelligence Index. The report covers the release, architecture and parameter counts, input and reasoning capabilities, the training method and the benchmark score.
Written by AI from the coverage · updated 3 hours ago
Coverage
Follow the reports to see the story from different sides.
- Hacker News front pageInkling 2 Small
Inkling 2 Small 发布，采用 MoE 架构，总参数量为 594B、激活参数量为 17B，支持图像输入与可调推理投入。模型在 Artificial Analysis Intelligence Index 上得分为 31，训练采用大规模统一 RL，覆盖编码、视觉、知识工作与推理任务。
Heat over time
Heat now 9·Comparable peak 10（Nov 6 09:00）·Comparable change over 24 hours –
The trend only compares accounts observed without gaps, so its range may be smaller than the current heat. Hover or tap the chart for each hour; the left and right arrow keys step through it.