Inkling 2 Small beats its own researchers at blind wine tasting using only a photo
Thinking Machines' second model generation is a 594B-parameter Mixture-of-Experts with 17B active parameters, scoring 31 on the Artificial Analysis Intelligence Index. It handles image inputs and variable effort, and shines at long-horizon autonomous work: it built a procedural Three.js city simulator, autonomously trained Qwen3.5-4B to forecast time series for $426 in 4.5 hours, and in a blind wine tasting identified three wines from a single photo in 19 of 20 trials. Weights arrive next week under Apache 2.0.
We even tested Inkling 2 Small against some of our researchers in a wine tasting event. It outperformed them using vision alone, correctly identifying three wines from a single picture in 19 out of 20 trials.
- Jackobrien
The best open models today have been “good enough” to do most knowledge work for a few months. This might be the first US one good enough too. That’s a big deal.
Open weight US based Opus 5.5 within 12 months.
- epsteingpt
It's a nice, domestic open weight model.
Without hundreds of millions in 9 days a small team is producing a model that can beat DeepSeek 4.1 and a platform to tune it.
Optimistic for the future.
- itkovian_
The most important thing here is training was entirely nvfp4 - this is best model by far that has ever been trained in that precision and explains why they were able to do it so fast
- a11r
Very refreshing to see benchmarks where the reporting team is not the clear winner. They are claiming to be better than GLM 5.3 flash, which is impressive if it holds up outside of benchmarks.
- tcdent
Is it SOTA yet? No. Are the domestic open weights labs demonstrating incredible velocity? Undeniably.