Hy4 is the stronger all-rounder.
Scores updated · 14 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.
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 Hy4 pulls ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+14.4points ahead
- Long, multi-file coding tasks in real codebasesSWE-bench Pro+8.2points ahead
- Long professional tasks in banking, consulting and lawAA ApexAgents+3.3points ahead
Where Seed 2.1 Pro Preview pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 14 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingHy4
- Terminal-Bench 2.1Hy4 by 14.485.471+14.4
- LMArena · WebDevHy4 by 112 rating points16321520+112 rating
- SWE-bench ProHy4 by 8.265.757.5+8.2
Other results9 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.
- MathArena Apex (Pass@1)Hy4 by 42.974.231.3+42.9
- Program BenchSeed 2.1 Pro Preview by 32.817.550.3+32.8
- DeepSWEHy4 by 31.664.332.7+31.6
- ToolathlonHy4 by 23.574.150.6+23.5
- PostTrainBenchHy4 by 19.135.616.5+19.1
- Agents' Last ExamSeed 2.1 Pro Preview by 18.622.841.4+18.6
- NL2RepoHy4 by 11.958.947+11.9
- CyberGymHy4 by 8.278.470.2+8.2
- OfficeQA ProSeed 2.1 Pro Preview by 666.272.2+6
Questions people ask
Which is better, Hy4 or Seed 2.1 Pro Preview?
Hy4 wins all two areas where both have results: coding and agents. Seed 2.1 Pro Preview wins none.
Which is better for coding?
Hy4. It wins 3 of the 3 coding tests both models report; Seed 2.1 Pro Preview wins none.
How do you compare the two?
We use the 14 benchmark tests both models have published scores on. The verdict counts the 5 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 9 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.