Agnes 2.5 Pro Beta vs Motif-3
Wins 3 of 5 areas
Agents · Reasoning · Long documents
Wins 1 of 5 areas
Facts
Agnes 2.5 Pro Beta is the stronger all-rounder.Motif-3 is better at facts.
Scores updated · 14 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.
- ReasoningHard problems that need careful thinking20Agnes 2.5 Pro Beta2 of 3 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own10Agnes 2.5 Pro Beta1 of 2 tests · 1 tie
- Long documentsFinding answers in very long texts10Agnes 2.5 Pro Beta1 of 1 test
- FactsGetting facts right instead of making them up02Motif-32 of 2 tests
- CodingWriting and fixing software11Even1 each
Long documents rests on a single test.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Agnes 2.5 Pro Beta pulls ahead
- Real work tasks from 44 professionsGDPVal+9.5points ahead
- Research-level physics problemsCritPt+9.1points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+7.1points ahead
Where Motif-3 pulls ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+13.3points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+5.2points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+4.4points ahead
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.
CodingEven
- SciCodeAgnes 2.5 Pro Beta by 6.648.842.2+6.6
- Terminal-Bench 2.1Motif-3 by 5.269.774.9+5.2
AgentsAgnes 2.5 Pro Beta
- GDPValAgnes 2.5 Pro Beta by 9.540.631.1+9.5
- τ-Bench V3 · Bankingtie35.735.3tie
ReasoningAgnes 2.5 Pro Beta
- CritPtAgnes 2.5 Pro Beta by 9.115.76.6+9.1
- GPQA DiamondAgnes 2.5 Pro Beta by 7.190.583.4+7.1
- Humanity's Last Examtie37.537tie
FactsMotif-3
- AA-Omniscience · AccuracyMotif-3 by 13.316.830.1+13.3
- AA-Omniscience · Non-hallucinationMotif-3 by 4.467.271.6+4.4
Other results4 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-OmniscienceMotif-3 by 20.7-10.510.2+20.7
- AA Agentic IndexAgnes 2.5 Pro Beta by 6.243.837.6+6.2
- AA IntelligenceAgnes 2.5 Pro Beta by 1.635.233.6+1.6
- Artificial Analysis Coding IndexMotif-3 by 1.262.363.5+1.2
Questions people ask
Which is better, Agnes 2.5 Pro Beta or Motif-3?
Agnes 2.5 Pro Beta wins three of the five areas where both have results: agents, reasoning and long documents. Motif-3 wins facts. They are level on coding.
Which is better for coding?
Neither. They win 1 coding test each of the 2 both models report.
How do you compare the two?
We use the 14 benchmark tests both models have published scores on. The verdict counts the 10 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 4 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.