Kimi K3 wins more areas, narrowly.DeepSeek-V4-Flash-Vision-Exp is cheaper and better at math.
Scores updated · 16 tests both models report · How we compare
Where each one wins
Tests won in each of the four 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.
Reasoning, math 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.
DeepSeek-V4-Flash-Vision-Exp costs 95% 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 K3 pulls ahead
- Programming problems, refreshed regularlyLiveBench · Coding+13.3points ahead
- Fresh logic puzzles that change regularlyLiveBench · Reasoning+5.3points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+1.1points ahead
Where DeepSeek-V4-Flash-Vision-Exp pulls ahead
- Fresh competition math problemsLiveBench · Mathematics+3.4points ahead
- Fixes real GitHub issues, working as an agentLiveBench · Agentic Coding+2.9points ahead
Every test, side by side
All 16 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingKimi K3
- LiveBench · CodingKimi K3 by 13.368.281.5+13.3
- LiveBench · Agentic CodingDeepSeek-V4-Flash-Vision-Exp by 2.965.162.2+2.9
- Terminal-Bench 2.1Kimi K3 by 1.183.985+1.1
MathDeepSeek-V4-Flash-Vision-Exp
- LiveBench · MathematicsDeepSeek-V4-Flash-Vision-Exp by 3.487.884.4+3.4
Following instructionsEven
- LiveBench · Instruction Followingtie7171.4tie
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.
- DeepSWEKimi K3 by 8.259.367.5+8.2
- ZeroBenchKimi K3 by 63541+6
- AutomationBench PublicKimi K3 by 5.125.730.8+5.1
- livebench_languageKimi K3 by 5.180.485.5+5.1
- CyberGymKimi K3 by 4.775.380+4.7
- Agents' Last ExamKimi K3 by 127.328.3+1
- livebench_data_analysistie79.578.7tie
- DSBench-Hardtie63.663tie
- Toolathlon Verifiedtie75.976.5tie
- NL2Repotie57.758tie
Questions people ask
Which is better, DeepSeek-V4-Flash-Vision-Exp or Kimi K3?
Kimi K3 wins two of the four areas where both have results: coding and reasoning. DeepSeek-V4-Flash-Vision-Exp wins math, and costs 95% less. They are level on following instructions.
Which is better for coding?
Kimi K3. It wins 2 of the 3 coding tests both models report; DeepSeek-V4-Flash-Vision-Exp wins 1.
Which is cheaper?
DeepSeek-V4-Flash-Vision-Exp costs $0.22 per million input tokens and $0.65 per million output tokens; Kimi K3 costs $3.00 and $15.00. That makes DeepSeek-V4-Flash-Vision-Exp about 95% cheaper for the same work.
How do you compare the two?
We use the 16 benchmark tests both models have published scores on. The verdict counts the 6 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.