What people said
Discussion collected automatically from Hacker News. We store short excerpts and link to the threads; we do not mirror them.
Community discussion around Inkling 2 Small has been mixed. The launch thread on Hacker News drew 274 points and 89 comments, with recurring themes being the wait for open weights, the model's size relative to consumer hardware, and how its benchmarks compare with Chinese open-weights models like DeepSeek and Kimi.
Threads
Themes
| Theme | Stance | Evidence |
|---|---|---|
| Price and cost efficiency | mixed | 14 of 40 sampled comments touch on this. Would this have been SOTA at the beginning of the year? Yes. Where you happy with SOTA in january? Why can't we then be happy with this? I mean, perhaps Inkling 3 or 4 Small will c |
| Open weights timing | positive | 8 of 40 sampled comments touch on this. Is it SOTA yet? No. Are the domestic open weights labs demonstrating incredible velocity? Undeniably. |
| Speed and latency | negative | 1 of 40 sampled comments touch on this. 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 fa |
| Benchmark skepticism | mixed | 4 of 40 sampled comments touch on this. 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 |
| Naming: why “Small”? | mixed | 9 of 40 sampled comments touch on this. Would this have been SOTA at the beginning of the year? Yes. Where you happy with SOTA in january? Why can't we then be happy with this? I mean, perhaps Inkling 3 or 4 Small will c |
In their words
Is it SOTA yet? No. Are the domestic open weights labs demonstrating incredible velocity? Undeniably.
Distillation is the tide that lifts all boats
Seems to be pretty close to SOTA on design. I’m excited to try it out with my agentic design system.
Would this have been SOTA at the beginning of the year? Yes. Where you happy with SOTA in january? Why can't we then be happy with this? I mean, perhaps Inkling 3 or 4 Small will crush Opus 5.5 and then we will ask if it matters since it's not Opus 7... Everyone can't be SOTA. With these open weight models, I simply ask. Is it useable, is it useful? Can I get very useful work out of it at a reasonable cost to other o
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.
> Open weight US based Opus 5.5 within 12 months. Seems unlikely. Maybe open weight at all near there? That'd be great.
Inception Mercury 3 and Particle-1's model are also good. (I am assuming Inception is an American company, not sure TBH)
Muse Glimmer is pretty good.
How this page works
We query the Hacker News search API for mentions of the model, keep the story metadata, and pull the comment tree for the largest thread only. Comments are trimmed to short excerpts and attributed to their authors, with a link back to the thread — the discussion belongs to the people who wrote it.