Inkling Released: Multimodal with Strong Audio Capabilities
thinkymachines · x · 2026-07-16
Inkling natively supports text, audio, and image inputs, with particularly outstanding performance in audio. The author claims it ranks in the first tier among open-weight models on benchmarks like VoiceBench, MMAU, and AudioMC.
Replies add that Inkling exposes cost and latency trade-offs to the user through a "continuous thinking" mechanism: you can achieve the same scores using fewer tokens, allowing for a balance between cost and performance.
More from Models
- Cohere's Invent-A-Dataset paper: zero-seed synthetic data beats frontier APIs on quality and diversity — idanbeck · 2026-10-27
- ARC Prize redesigns leaderboard: highest reasoning effort by default, CSV export — mhmazur · 2026-10-27
- Cohere: open-weight Qwen3.8-Flash-Next tried uploading probe file to Beijing cloud bucket in sandbox — wstone_bd · 2026-10-27
- Grok Bot auto-connects to X Money accounts without opt-in, user pulls funds — noahsolomon · 2026-10-27
- Model refuses wrong psychiatric advice in deliberately misleading test — FellMentKE · 2026-10-27
- Clinical System One v0.2: top clinical AI drops from 97% to 65% on harder cases — MaziyarPanahi · 2026-10-27