Gemini 3.1 Flash Lite Preview vs Kimi K2.5
Wins 2 of 7 areas
Facts · Following instructions
Wins 5 of 7 areas
Coding · Agents · Reasoning · Images and charts · Long documents
Kimi K2.5 is the stronger all-rounder.Gemini 3.1 Flash Lite Preview is cheaper and better at facts.
Scores updated · 29 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.
- CodingWriting and fixing software04Kimi K2.54 of 4 tests
- ReasoningHard problems that need careful thinking03Kimi K2.53 of 3 tests
- AgentsCarrying out multi-step tasks on its own02Kimi K2.52 of 3 tests · 1 tie
- Images and chartsUnderstanding pictures, charts and video02Kimi K2.52 of 3 tests · 1 tie
- Long documentsFinding answers in very long texts01Kimi K2.51 of 1 test
- FactsGetting facts right instead of making them up21Gemini 3.1 Flash Lite Preview2 of 3 tests
- Following instructionsDoing exactly what it is asked10Gemini 3.1 Flash Lite Preview1 of 2 tests · 1 tie
Long documents rests on a single test.
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 3.1 Flash Lite Preview costs 51% 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 Kimi K2.5 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+17points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+14.6points ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+13.5points ahead
Where Gemini 3.1 Flash Lite Preview pulls ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+7points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+1.1points ahead
Every test, side by side
All 29 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingKimi K2.5
- LMArena · WebDevKimi K2.5 by 182 rating points12541436+182 rating
- Terminal-Bench 2.1Kimi K2.5 by 14.631.145.7+14.6
- Terminal-Bench HardKimi K2.5 by 10.624.234.8+10.6
- SciCodeKimi K2.5 by 5.643.449+5.6
AgentsKimi K2.5
- GDPValKimi K2.5 by 17.2017.2+17.2
- τ-Bench V3 · BankingKimi K2.5 by 4.59.714.2+4.5
- AA ApexAgentstie12.211.5tie
ReasoningKimi K2.5
- Humanity's Last ExamKimi K2.5 by 13.517.230.7+13.5
- GPQA DiamondKimi K2.5 by 5.782.287.9+5.7
- CritPtKimi K2.5 by 21.13.1+2
FactsGemini 3.1 Flash Lite Preview
- AA-Omniscience · Non-hallucinationKimi K2.5 by 1717.334.3+17
- Vectara HHEM hallucination ratelower is betterGemini 3.1 Flash Lite Preview by 68.214.2+6
- AA-Omniscience · AccuracyGemini 3.1 Flash Lite Preview by 1.136.335.2+1.1
Images and chartsKimi K2.5
- CharXiv (RQ)Kimi K2.5 by 4.373.277.5+4.3
- LMArena · VisionKimi K2.5 by 29 rating points12401269+29 rating
- MMMU-Protie75.575.4tie
Following instructionsGemini 3.1 Flash Lite Preview
- IFBenchGemini 3.1 Flash Lite Preview by 777.270.2+7
- Multi-Challengetie60.661.4tie
Other results10 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.
- τ²-Bench Telecom (AA run)Kimi K2.5 by 64.631.395.9+64.6
- GDPval-AA v2Kimi K2.5 by 367 rating points6421009+367 rating
- AA Agentic IndexKimi K2.5 by 18.53.221.7+18.5
- Artificial Analysis Coding IndexKimi K2.5 by 12.134.746.8+12.1
- AA-OmniscienceKimi K2.5 by 9.1-16.4-7.3+9.1
- AA IntelligenceKimi K2.5 by 7.915.623.5+7.9
- vectara_answer_rateGemini 3.1 Flash Lite Preview by 7.499.692.2+7.4
- vectara_factual_consistencyGemini 3.1 Flash Lite Preview by 691.885.8+6
- vectara_avg_summary_lengthKimi K2.5 by 49.4 rating points62.6112+49.4 rating
- VideoMMMUKimi K2.5 by 1.884.886.6+1.8
Questions people ask
Which is better, Gemini 3.1 Flash Lite Preview or Kimi K2.5?
Kimi K2.5 wins five of the seven areas where both have results: coding, agents, reasoning, images and charts and long documents. Gemini 3.1 Flash Lite Preview wins facts and following instructions, and costs 51% less.
Which is better for coding?
Kimi K2.5. It wins 4 of the 4 coding tests both models report; Gemini 3.1 Flash Lite Preview wins none.
Which is cheaper?
Gemini 3.1 Flash Lite Preview costs $0.25 per million input tokens and $1.50 per million output tokens; Kimi K2.5 costs $0.60 and $3.00. That makes Gemini 3.1 Flash Lite Preview about 51% cheaper for the same work.
How do you compare the two?
We use the 29 benchmark tests both models have published scores on. The verdict counts the 19 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 10 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.