Gemma 4 31B vs Kimi K3
Wins 0 of 7 areas
—
Wins 7 of 7 areas
Coding · Agents · Reasoning · Facts · Images and charts · Math · Long documents
Kimi K3 is the stronger all-rounder.Gemma 4 31B is cheaper.
Scores updated · 39 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.
- AgentsCarrying out multi-step tasks on its own06Kimi K36 of 6 tests
- CodingWriting and fixing software03Kimi K33 of 3 tests
- ReasoningHard problems that need careful thinking03Kimi K33 of 3 tests
- FactsGetting facts right instead of making them up03Kimi K33 of 3 tests
- Images and chartsUnderstanding pictures, charts and video02Kimi K32 of 2 tests
- MathCompetition and research-level math02Kimi K32 of 2 tests
- Long documentsFinding answers in very long texts02Kimi K32 of 2 tests
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.
Gemma 4 31B costs 97% 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
- Finds hard-to-locate facts by browsing the webBrowseComp+54.2points ahead
- Real work tasks from 44 professionsGDPVal+45.7points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+41.6points ahead
Where Gemma 4 31B pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 39 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingKimi K3
- Terminal-Bench 2.1Kimi K3 by 41.643.485+41.6
- LMArena · WebDevKimi K3 by 294 rating points13651659+294 rating
- SciCodeKimi K3 by 1445.559.5+14
AgentsKimi K3
- BrowseCompKimi K3 by 54.23791.2+54.2
- GDPValKimi K3 by 45.7651.7+45.7
- τ-Bench V3 · BankingKimi K3 by 31.214.846+31.2
- MCP AtlasKimi K3 by 3054.284.2+30
- Terminal-Bench 4.0Kimi K3 by 12.6012.6+12.6
- AA IT-Bench SREKimi K3 by 10.437.347.7+10.4
ReasoningKimi K3
- Humanity's Last ExamKimi K3 by 23.323.646.9+23.3
- CritPtKimi K3 by 221.423.4+22
- GPQA DiamondKimi K3 by 7.885.793.5+7.8
FactsKimi K3
- SimpleQA VerifiedKimi K3 by 40.210.450.6+40.2
- AA-Omniscience · Non-hallucinationKimi K3 by 31.81546.8+31.8
- AA-Omniscience · AccuracyKimi K3 by 27.62047.6+27.6
Images and chartsKimi K3
- CharXiv (RQ)Kimi K3 by 23.467.991.3+23.4
- MMMU-ProKimi K3 by 7.173.480.5+7.1
MathKimi K3
- HMMT Feb. 2026Kimi K3 by 19.877.297+19.8
- AIME 2026Kimi K3 by 7.589.296.7+7.5
Long documentsKimi K3
- AA-LCRKimi K3 by 1969.788.7+19
- LongBench v2Kimi K3 by 16.152.969+16.1
Other results18 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.
- GDPval-AA 2.1Kimi K3 by 916 rating points6271543+916 rating
- SWEBench Pro v2 (Hard)Kimi K3 by 7414.288.2+74
- ScreenSpot Pro originalKimi K3 by 67.95.973.8+67.9
- AA-OmniscienceKimi K3 by 67.6-47.919.7+67.6
- JobBenchKimi K3 by 44.78.252.9+44.7
- AA Agentic IndexKimi K3 by 43.96.750.6+43.9
- NL2RepoKimi K3 by 42.515.558+42.5
- CritPt v2 HardKimi K3 by 36.5440.5+36.5
- DeepSearchQA (F1)Kimi K3 by 33.361.795+33.3
- HLE (with tools)Kimi K3 by 33.326.559.8+33.3
- Artificial Analysis Coding IndexKimi K3 by 32.843.476.2+32.8
- AA IntelligenceKimi K3 by 28.914.743.6+28.9
- HLE-DiamondKimi K3 by 19.77.226.9+19.7
- GDP (Surge AI)Kimi K3 by 16622+16
- Chartography originalKimi K3 by 14.61226.6+14.6
- Terminal-Bench 4.0Kimi K3 by 12.6012.6+12.6
- MATH-VisionKimi K3 by 12.285.697.8+12.2
- Terminal-Bench-Science 0.1Kimi K3 by 8.108.1+8.1
Questions people ask
Which is better, Gemma 4 31B or Kimi K3?
Kimi K3 wins all seven areas where both have results: coding, agents, reasoning, facts, images and charts, math and long documents. Gemma 4 31B wins none, but costs 97% less.
Which is better for coding?
Kimi K3. It wins 3 of the 3 coding tests both models report; Gemma 4 31B wins none.
Which is cheaper?
Gemma 4 31B costs $0.15 per million input tokens and $0.40 per million output tokens; Kimi K3 costs $3.00 and $15.00. That makes Gemma 4 31B about 97% cheaper for the same work.
How do you compare the two?
We use the 39 benchmark tests both models have published scores on. The verdict counts the 21 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 18 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.