Atria Dawn is the stronger all-rounder.
Scores updated · 5 tests both models report · How we compare
Where each one wins
Tests won in each of the two areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
Agents rests on a single test.
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Atria Dawn pulls ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+13.9points ahead
- Finds hard-to-locate facts by browsing the webBrowseComp+8.3points ahead
- Long, multi-file coding tasks in real codebasesSWE-bench Pro+1.7points ahead
Where Hy3-preview pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 5 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingAtria Dawn
- Terminal-Bench 2.1Atria Dawn by 13.978.364.4+13.9
- SWE-bench ProAtria Dawn by 1.759.657.9+1.7
Other results2 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- WideSearchAtria Dawn by 5.581.976.4+5.5
- DeepSearchQA (F1)Atria Dawn by 59691+5
Questions people ask
Which is better, Atria Dawn or Hy3-preview?
Atria Dawn wins all two areas where both have results: coding and agents. Hy3-preview wins none.
Which is better for coding?
Atria Dawn. It wins 2 of the 2 coding tests both models report; Hy3-preview wins none.
How do you compare the two?
We use the 5 benchmark tests both models have published scores on. The verdict counts the 3 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 2 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.