Inkling from @thinkymachines is live on OpenRouter. An open-weights MoE model with 975B total / 41B active parameters, 1M context, and controllable reasoning across text, images, and audio. Try it:

Inkling running on a DGX B200 node with vLLM across 8 GPUs. Text, images, and reasoning from a single container. Red Hat AI FP8 checkpoints coming soon. Shoutout to the @vllm_project community for getting this up and running and @_soyr_ for the quick video.

BREAKING: Inkling by @thinkymachines is 9th overall on Agentic Web App Arena by Design Arena with an Elo of 1257 It's an open-weight model in the same performance band as Claude Opus 4.6 by @AnthropicAI and Gemini 3.5 Flash by @GoogleDeepMind This makes Inkling the highest-ranking US-based open-weight model for agentic workloads, achieving frontier-level performance Congrats to the @thinkymachines team for this achievement!

We want to improve Inkling’s agentic performance. To help us understand its real-world behavior, we are making it available for free on OpenRouter (only with agentic harnesses) for the next few weeks, starting now. We’ll use the data, disassociated from accounts, to better it.

Thinking Machines just released Inkling-Small: 276B total, 12B active. A faster Inkling that matches or beats its 975B big brother in many benchmarks. To test its speed, we plugged it into HF's speech-to-speech. Audio goes directly into the model, and it replies using faster-Qwen3TTS. Running on 8x RTX Pro 6000 Blackwell, we get audio back in under 500ms. The normal Inkling needs 2TB of VRAM, this one fits on one node. Because the model hears the audio instead of a transcript, it can hear your tone and emotions. You can check it in the video. Or just go and try it in the space! Really fun model:

Congrats to @thinkymachines on the release of Inkling-small! A smaller variant of Inkling, now live on our AudioMultiChallenge and MCP Atlas leaderboards. Inkling-small is tied for🥇on AudioMultiChallenge, scoring about the same as the larger Inkling despite the size difference. Strong multi-turn audio performance has typically come from the biggest models, so this is a promising signal for teams building voice applications where latency and cost matter. Also notable, on MCP Atlas it ranks second among open models on tool calling, behind Kimi K3 and ahead of GLM 5.2. Holding up on both audio reasoning and tool use at this size is a strong showing.

Thinking Machines just dropped the best open weight AI model outside of China! Inkling beats Nemotron 3 Ultra and benchmarks put it between Kimi 2.5 & 2.6. Many were contending to this throne, but Thinky has come out on top. Really solid release, and will pair well with Tinker.