GLM-4.5-Air vs Kimi K2 Instruct
Wins 0 of 6 areas
—
Wins 5 of 6 areas
Coding · Reasoning · Facts · Long documents · Following instructions
Kimi K2 Instruct is the stronger all-rounder.GLM-4.5-Air is cheaper.
Scores updated · 31 tests both models report · How we compare
Where each one wins
Tests won in each of the six 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 software02Kimi K2 Instruct2 of 3 tests · 1 tie
- ReasoningHard problems that need careful thinking01Kimi K2 Instruct1 of 3 tests · 2 ties
- FactsGetting facts right instead of making them up12Kimi K2 Instruct2 of 3 tests
- Long documentsFinding answers in very long texts01Kimi K2 Instruct1 of 1 test
- Following instructionsDoing exactly what it is asked01Kimi K2 Instruct1 of 1 test
- AgentsCarrying out multi-step tasks on its own11Even1 each
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.
GLM-4.5-Air costs 63% 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 Instruct pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+23.2points ahead
- Real work tasks from 44 professionsGDPVal+15.2points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+9.1points ahead
Where GLM-4.5-Air pulls ahead
- Finds hard-to-locate facts by browsing the webBrowseComp+7.2points ahead
Every test, side by side
All 31 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingKimi K2 Instruct
- SWE-bench VerifiedKimi K2 Instruct by 8.257.665.8+8.2
- Terminal-Bench HardKimi K2 Instruct by 320.523.5+3
- SciCodetie30.630.7tie
AgentsEven
- GDPValKimi K2 Instruct by 15.2318.2+15.2
- BrowseCompGLM-4.5-Air by 7.221.314.1+7.2
ReasoningKimi K2 Instruct
- GPQA DiamondKimi K2 Instruct by 3.473.376.7+3.4
- Humanity's Last Examtie76.4tie
- CritPttie00tie
FactsKimi K2 Instruct
- AA-Omniscience · Non-hallucinationKimi K2 Instruct by 23.27.130.3+23.2
- AA-Omniscience · AccuracyKimi K2 Instruct by 9.116.325.4+9.1
- Vectara HHEM hallucination ratelower is betterGLM-4.5-Air by 8.69.317.9+8.6
Following instructionsKimi K2 Instruct
- IFBenchKimi K2 Instruct by 4.137.641.7+4.1
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.
- HarmBenchKimi K2 Instruct by 41.256.197.4+41.2
- AA-OmniscienceKimi K2 Instruct by 34.9-61.5-26.6+34.9
- AIME 2025GLM-4.5-Air by 33.883.349.5+33.8
- τ²-Bench Telecom (AA run)Kimi K2 Instruct by 26.946.573.4+26.9
- AIR-Bench 2024Kimi K2 Instruct by 1757.174.1+17
- LiveCodeBenchGLM-4.5-Air by 1770.753.7+17
- AA Agentic IndexKimi K2 Instruct by 16.72137.7+16.7
- vectara_avg_summary_lengthGLM-4.5-Air by 11.470.659.2+11.4
- vectara_factual_consistencyGLM-4.5-Air by 8.690.782.1+8.6
- AA IntelligenceKimi K2 Instruct by 4.211.115.3+4.2
- Artificial Analysis Coding IndexKimi K2 Instruct by 2.123.825.9+2.1
- simple_safety_testsKimi K2 Instruct by 199100+1
- MATH-500 (EM)tie98.197.4tie
- vectara_answer_ratetie98.198.6tie
- XSTesttie98.698.2tie
- anthropic_red_teamtie9999.3tie
- MMLU-Protie81.481.1tie
- Terminal-Benchtie3030tie
Questions people ask
Which is better, GLM-4.5-Air or Kimi K2 Instruct?
Kimi K2 Instruct wins five of the six areas where both have results: coding, reasoning, facts, long documents and following instructions. GLM-4.5-Air wins none, but costs 63% less. They are level on agents.
Which is better for coding?
Kimi K2 Instruct. It wins 2 of the 3 coding tests both models report; GLM-4.5-Air wins none, and 1 is a tie.
Which is cheaper?
GLM-4.5-Air costs $0.17 per million input tokens and $0.98 per million output tokens; Kimi K2 Instruct costs $0.60 and $2.50. That makes GLM-4.5-Air about 63% cheaper for the same work.
How do you compare the two?
We use the 31 benchmark tests both models have published scores on. The verdict counts the 13 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.