Gemini 2.5 Flash vs Gemini 2.5 Pro
Wins 2 of 7 areas
Agents · Following instructions
Wins 4 of 7 areas
Coding · Reasoning · Images and charts · Long documents
Gemini 2.5 Pro wins more areas, narrowly.Gemini 2.5 Flash is cheaper and better at agents.
Scores updated · 45 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 thinking05Gemini 2.5 Pro5 of 5 tests
- CodingWriting and fixing software03Gemini 2.5 Pro3 of 3 tests
- Images and chartsUnderstanding pictures, charts and video02Gemini 2.5 Pro2 of 3 tests · 1 tie
- Long documentsFinding answers in very long texts01Gemini 2.5 Pro1 of 1 test
- AgentsCarrying out multi-step tasks on its own10Gemini 2.5 Flash1 of 1 test
- Following instructionsDoing exactly what it is asked10Gemini 2.5 Flash1 of 1 test
- FactsGetting facts right instead of making them up11Even1 each · 1 tie
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 costs 75% less for the same work.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Gemini 2.5 Pro pulls ahead
- Code for real scientific research problemsSciCode+17.2points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+13.1points ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+12.9points ahead
Where Gemini 2.5 Flash pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+15.6points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+1.6points ahead
Every test, side by side
All 45 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGemini 2.5 Pro
- SciCodeGemini 2.5 Pro by 17.229.146.3+17.2
- Terminal-Bench HardGemini 2.5 Pro by 12.913.626.5+12.9
- SWE-bench VerifiedGemini 2.5 Pro by 2.860.463.2+2.8
ReasoningGemini 2.5 Pro
- SimpleBenchGemini 2.5 Pro by 10.441.251.6+10.4
- Humanity's Last ExamGemini 2.5 Pro by 5.912.118+5.9
- GPQA DiamondGemini 2.5 Pro by 4.67983.6+4.6
- ARC-AGI-2Gemini 2.5 Pro by 2.42.54.9+2.4
- CritPtGemini 2.5 Pro by 1.51.12.6+1.5
FactsEven
- AA-Omniscience · Non-hallucinationGemini 2.5 Flash by 15.624.79.1+15.6
- AA-Omniscience · AccuracyGemini 2.5 Pro by 13.125.939+13.1
- Vectara HHEM hallucination ratelower is bettertie7.87tie
Images and chartsGemini 2.5 Pro
- MMMU-ProGemini 2.5 Pro by 5.869.174.9+5.8
- LMArena · VisionGemini 2.5 Pro by 28 rating points12351263+28 rating
- MMMUtie79.779.6tie
Following instructionsGemini 2.5 Flash
- IFBenchGemini 2.5 Flash by 1.650.348.7+1.6
Other results28 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.
- MRCRGemini 2.5 Pro by 613293+61
- Artificial Analysis Coding IndexGemini 2.5 Pro by 24.522.246.7+24.5
- SimpleQAGemini 2.5 Pro by 23.926.950.8+23.9
- τ²-Bench Telecom (AA run)Gemini 2.5 Pro by 22.531.654.1+22.5
- HMMT Feb. 2025Gemini 2.5 Pro by 16.664.280.8+16.6
- AA Agentic IndexGemini 2.5 Flash by 15.318.83.5+15.3
- Aider-PolyglotGemini 2.5 Pro by 14.661.976.5+14.6
- AA-OmniscienceGemini 2.5 Pro by 13.5-29.8-16.4+13.5
- LiveCodeBench (v5)Gemini 2.5 Pro by 11.763.975.6+11.7
- livecodebench_hardGemini 2.5 Pro by 11.148.359.4+11.1
- AIME 2025Gemini 2.5 Pro by 117283+11
- AIME 2024Gemini 2.5 Pro by 9.782.392+9.7
- livecodebench_mediumGemini 2.5 Pro by 8.182.590.6+8.1
- AIR-Bench 2024Gemini 2.5 Pro by 4.968.773.6+4.9
- ARC-AGI-1Gemini 2.5 Pro by 4.732.337+4.7
- HarmBenchGemini 2.5 Pro by 2.862.665.4+2.8
- LiveCodeBenchGemini 2.5 Flash by 276.274.2+2
- AA IntelligenceGemini 2.5 Pro by 1.913.115+1.9
- bbqGemini 2.5 Flash by 1.397.796.4+1.3
- simple_safety_testsGemini 2.5 Flash by 19897+1
- vectara_factual_consistencytie92.293tie
- anthropic_red_teamtie98.899.5tie
- vectara_avg_summary_lengthtie101.5106.4tie
- livecodebench_easytie99.198.8tie
- Global-MMLU-Litetie88.488.6tie
- vectara_answer_ratetie9999.1tie
- XSTesttie98.898.7tie
- frontiermath_tier_4_v1tie4.24.2tie
Questions people ask
Which is better, Gemini 2.5 Flash or Gemini 2.5 Pro?
Gemini 2.5 Pro wins four of the seven areas where both have results: coding, reasoning, images and charts and long documents. Gemini 2.5 Flash wins agents and following instructions, and costs 75% less. They are level on facts.
Which is better for coding?
Gemini 2.5 Pro. It wins 3 of the 3 coding tests both models report; Gemini 2.5 Flash wins none.
Which is cheaper?
Gemini 2.5 Flash costs $0.30 per million input tokens and $2.50 per million output tokens; Gemini 2.5 Pro costs $1.25 and $10.00. That makes Gemini 2.5 Flash about 75% cheaper for the same work.
How do you compare the two?
We use the 45 benchmark tests both models have published scores on. The verdict counts the 17 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 28 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.