Agnes 3.0 Flash vs mercury-2.5-preview
Wins 5 of 5 areas
Coding · Agents · Reasoning · Facts · Long documents
Wins 0 of 5 areas
—
Agnes 3.0 Flash is the stronger all-rounder.
Scores updated · 10 tests both models report · How we compare
Where each one wins
Tests won in each of the five 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.
- AgentsCarrying out multi-step tasks on its own20Agnes 3.0 Flash2 of 2 tests
- ReasoningHard problems that need careful thinking20Agnes 3.0 Flash2 of 2 tests
- FactsGetting facts right instead of making them up20Agnes 3.0 Flash2 of 2 tests
- CodingWriting and fixing software10Agnes 3.0 Flash1 of 1 test
- Long documentsFinding answers in very long texts10Agnes 3.0 Flash1 of 1 test
Coding and long documents rest on a single test each.
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 Agnes 3.0 Flash pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+31.9points ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+26.7points ahead
- Code for real scientific research problemsSciCode+13.1points ahead
Where mercury-2.5-preview pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 10 tests both models report. The winning score is in its model's colour; marks a score checked independently.
AgentsAgnes 3.0 Flash
- GDPValAgnes 3.0 Flash by 52.552.50+52.5
- Terminal-Bench 4.0Agnes 3.0 Flash by 7.17.10+7.1
ReasoningAgnes 3.0 Flash
- Humanity's Last ExamAgnes 3.0 Flash by 26.738.511.8+26.7
- CritPtAgnes 3.0 Flash by 15.115.10+15.1
FactsAgnes 3.0 Flash
- AA-Omniscience · Non-hallucinationAgnes 3.0 Flash by 31.951.619.7+31.9
- AA-Omniscience · AccuracyAgnes 3.0 Flash by 2.825.522.6+2.8
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.
- AA-OmniscienceAgnes 3.0 Flash by 28.9-10.6-39.5+28.9
- AA IntelligenceAgnes 3.0 Flash by 23.235.512.3+23.2
Questions people ask
Which is better, Agnes 3.0 Flash or mercury-2.5-preview?
Agnes 3.0 Flash wins all five areas where both have results: coding, agents, reasoning, facts and long documents. mercury-2.5-preview wins none.
Which is better for coding?
Agnes 3.0 Flash. It wins the one coding test both models report.
How do you compare the two?
We use the 10 benchmark tests both models have published scores on. The verdict counts the 8 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.