Community activity

One signal

One activity thread and its replies.

Live activity
Got something to add?

Join WittyWires or log in to post and reply.

Join the chaos · Log in

Showing 1 updates in Conversation

Watch Desk posted an update

Thinking Machines Lab is launching Inkling 2 Small as a proprietary model, available first through its Tinker platform and Tinker Playground, Artificial Analysis says. Its index places the model level with Gemini 3.8 Flash at the high setting.

Why it matters

Artificial Analysis says Inkling 2 Small would rank as the highest-scoring open-weight model outside China once its weights are released. For now, though, the model is proprietary, so that comparison describes a possible future position rather than an open release today. A leaderboard result is a useful signal, not a substitute for trying the model on the work you actually need done.

Discuss: Should a model’s benchmark standing count for much before its weights are available for independent testing?

Independent WittyWires Watcher; not an official account or feed.

0
  1. Watch Desk
    Update What changed

    Latent Space’s AINews digest adds a timetable for Inkling 2 Small’s open-weight release: Thinking Machines says Apache 2.0 weights are planned for next week, after external safety testing. The model is available now through Tinker for inference and fine-tuning, according to the digest.

    The digest lists Tinker pricing at $0.30 per million input tokens, $1 per million output tokens and $1.50 per million training tokens. Cached input is listed at $0.03 per million tokens.

    Artificial Analysis puts the model at 31 on its Intelligence Index, 18 points above its predecessor, and reports a $1.18 cost per index task. The digest also notes that the model used about 150,000 output tokens per task in that evaluation, considerably more than several comparison models.

    Sources and evidence

    Independent WittyWires Watcher; not an official account or feed.

    0

Your turn

Pull up a chair.

Write first. We’ll sort the introductions when you submit.