Claude 3.7 Sonnet vs Gemini 2.5 Flash Lite Preview 09-2025
Wins 4 of 7 areas
Coding · Agents · Reasoning · Facts
Wins 3 of 7 areas
Images and charts · Long documents · Following instructions
Claude 3.7 Sonnet wins more areas, narrowly.Gemini 2.5 Flash Lite Preview 09-2025 is cheaper and better at images and charts.
Scores updated · 17 tests both models report · How we compare
Where each one wins
Tests won in each of the seven 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 thinking20Claude 3.7 Sonnet2 of 3 tests · 1 tie
- CodingWriting and fixing software20Claude 3.7 Sonnet2 of 2 tests
- FactsGetting facts right instead of making them up20Claude 3.7 Sonnet2 of 2 tests
- AgentsCarrying out multi-step tasks on its own10Claude 3.7 Sonnet1 of 1 test
- Images and chartsUnderstanding pictures, charts and video02Gemini 2.5 Flash Lite Preview 09-20252 of 2 tests
- Long documentsFinding answers in very long texts01Gemini 2.5 Flash Lite Preview 09-20251 of 1 test
- Following instructionsDoing exactly what it is asked01Gemini 2.5 Flash Lite Preview 09-20251 of 1 test
Agents, long documents and following instructions 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.
Gemini 2.5 Flash Lite Preview 09-2025 costs 97% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Claude 3.7 Sonnet pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+47.8points ahead
- Code for real scientific research problemsSciCode+11.6points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+10points ahead
Every test, side by side
All 17 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude 3.7 Sonnet
- SciCodeClaude 3.7 Sonnet by 11.640.328.7+11.6
- Terminal-Bench HardClaude 3.7 Sonnet by 8.321.212.9+8.3
ReasoningClaude 3.7 Sonnet
- GPQA DiamondClaude 3.7 Sonnet by 6.377.270.9+6.3
- Humanity's Last ExamClaude 3.7 Sonnet by 2.79.77+2.7
- CritPttie0.90tie
FactsClaude 3.7 Sonnet
- AA-Omniscience · Non-hallucinationClaude 3.7 Sonnet by 47.86012.2+47.8
- AA-Omniscience · AccuracyClaude 3.7 Sonnet by 102818+10
Images and chartsGemini 2.5 Flash Lite Preview 09-2025
- MMMU-ProGemini 2.5 Flash Lite Preview 09-2025 by 4.960.165+4.9
- LMArena · VisionGemini 2.5 Flash Lite Preview 09-2025 by 29 rating points11691198+29 rating
Long documentsGemini 2.5 Flash Lite Preview 09-2025
- AA-LCRGemini 2.5 Flash Lite Preview 09-2025 by 1351.764.7+13
Following instructionsGemini 2.5 Flash Lite Preview 09-2025
- IFBenchGemini 2.5 Flash Lite Preview 09-2025 by 4.348.352.6+4.3
Other results5 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-OmniscienceClaude 3.7 Sonnet by 53.3-0.7-54+53.3
- τ²-Bench Telecom (AA run)Claude 3.7 Sonnet by 2454.730.7+24
- Artificial Analysis Coding IndexClaude 3.7 Sonnet by 14.336.422.1+14.3
- AA Agentic IndexClaude 3.7 Sonnet by 143723+14
- AA IntelligenceClaude 3.7 Sonnet by 7.317.710.4+7.3
Questions people ask
Which is better, Claude 3.7 Sonnet or Gemini 2.5 Flash Lite Preview 09-2025?
Claude 3.7 Sonnet wins four of the seven areas where both have results: coding, agents, reasoning and facts. Gemini 2.5 Flash Lite Preview 09-2025 wins images and charts, long documents and following instructions, and costs 97% less.
Which is better for coding?
Claude 3.7 Sonnet. It wins 2 of the 2 coding tests both models report; Gemini 2.5 Flash Lite Preview 09-2025 wins none.
Which is cheaper?
Claude 3.7 Sonnet costs $3.00 per million input tokens and $15.00 per million output tokens; Gemini 2.5 Flash Lite Preview 09-2025 costs $0.10 and $0.40. That makes Gemini 2.5 Flash Lite Preview 09-2025 about 97% cheaper for the same work.
How do you compare the two?
We use the 17 benchmark tests both models have published scores on. The verdict counts the 12 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 5 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.