Inkling Open-Weight Model Supports Audio Transcription
ziqiao_ma · x · 2026-07-19
According to a summary by Artificial Analysis, Thinking Machines has released a new open-weight model named Inkling. It features 975B total parameters and 41B active parameters, supporting text, image, and audio inputs for speech transcription.
Tested on the AA-WER speech recognition leaderboard with a 256K context version, it achieved a 3.5% word error rate, ranking second among open-weight models, just behind Mistral's Voxtral Small (2.8%). The post also mentions that Inkling's weights are openly available on Hugging Face under the Apache 2.0 license, making it currently the largest open-weight model supporting transcription on that leaderboard.
More from Models
- What are the best models to run on 48 GB of VRAM with two RTX 3090s? — ludos1978 · 2026-07-22
- Google releases Gemini 3.6 Flash as Gemini 3.5 Pro remains in testing — Ars Technica AI · 2026-07-22
- Google says Gemini 3.5 Pro is in partner testing as Gemini 4 pre-training starts — haider1 · 2026-07-22
- A benchmark chart puts a flash model around 5th place, but critics say it is far pricier — soumitrashukla9 · 2026-07-22
- Google introduces three new Gemini models focused on speed, token efficiency, and scale — Jason_perei · 2026-07-22
- How to Distinguish Genuine Token Efficiency from Shorter, Omissive Answers? — ruthstarkman · 2026-07-22