GLM-5.2 vs Kimi K2.7 Code
Wins 5 of 7 areas
Coding · Agents · Reasoning · Math · Following instructions
Wins 2 of 7 areas
Facts · Long documents
GLM-5.2 is the stronger all-rounder.Kimi K2.7 Code is cheaper and better at facts.
Scores updated · 34 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 software60GLM-5.26 of 6 tests
- MathCompetition and research-level math30GLM-5.23 of 3 tests
- AgentsCarrying out multi-step tasks on its own20GLM-5.22 of 4 tests · 2 ties
- Following instructionsDoing exactly what it is asked20GLM-5.22 of 2 tests
- ReasoningHard problems that need careful thinking21GLM-5.22 of 5 tests · 2 ties
- FactsGetting facts right instead of making them up12Kimi K2.7 Code2 of 3 tests
- Long documentsFinding answers in very long texts01Kimi K2.7 Code1 of 1 test
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.
Kimi K2.7 Code costs 15% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where GLM-5.2 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+56.1points ahead
- Extremely hard research-level math problemsFrontierMath Tier 4+17.1points ahead
- Real work tasks from 44 professionsGDPVal+16.7points ahead
Where Kimi K2.7 Code pulls ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+15.3points ahead
- Fresh logic puzzles that change regularlyLiveBench · Reasoning+4.2points ahead
- Short factual questions, answered correctlySimpleQA Verified+2.3points ahead
Every test, side by side
All 34 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGLM-5.2
- LMArena · WebDevGLM-5.2 by 130 rating points16031473+130 rating
- Terminal-Bench 2.1GLM-5.2 by 10.577.967.4+10.5
- LiveBench · Agentic CodingGLM-5.2 by 6.151.845.7+6.1
- Terminal-Bench HardGLM-5.2 by 6.150.844.7+6.1
- LiveBench · CodingGLM-5.2 by 5.779.774+5.7
- SciCodeGLM-5.2 by 3.451.247.8+3.4
AgentsGLM-5.2
- GDPValGLM-5.2 by 16.743.727+16.7
- τ-Bench V3 · BankingGLM-5.2 by 14.434.620.2+14.4
- MCP Atlastie76.876tie
- Terminal-Bench 4.0tie11tie
ReasoningGLM-5.2
- CritPtGLM-5.2 by 10.920.910+10.9
- Humanity's Last ExamGLM-5.2 by 6.141.135+6.1
- LiveBench · ReasoningKimi K2.7 Code by 4.278.682.8+4.2
- SimpleBenchtie58.857.9tie
- GPQA Diamondtie89.589.6tie
FactsKimi K2.7 Code
- AA-Omniscience · Non-hallucinationGLM-5.2 by 56.173.717.6+56.1
- AA-Omniscience · AccuracyKimi K2.7 Code by 15.324.339.6+15.3
- SimpleQA VerifiedKimi K2.7 Code by 2.334.236.5+2.3
MathGLM-5.2
- FrontierMath Tier 4GLM-5.2 by 17.129.312.2+17.1
- LiveBench · MathematicsGLM-5.2 by 10.289.879.6+10.2
- FrontierMath Tiers 1-3 (v2)GLM-5.2 by 5.259.254+5.2
Following instructionsGLM-5.2
- IFBenchGLM-5.2 by 10.273.363.1+10.2
- LiveBench · Instruction FollowingGLM-5.2 by 662.356.3+6
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.
- AA Agentic IndexGLM-5.2 by 16.939.422.5+16.9
- AA-OmniscienceGLM-5.2 by 14.64.4-10.2+14.6
- DeepSWE 1.1GLM-5.2 by 134431+13
- livebench_data_analysisGLM-5.2 by 1173.762.7+11
- Program BenchGLM-5.2 by 10.163.753.6+10.1
- τ²-Bench Telecom (AA run)GLM-5.2 by 999.190.1+9
- Artificial Analysis Coding IndexGLM-5.2 by 868.860.8+8
- AA IntelligenceGLM-5.2 by 7.933.725.8+7.9
- MLS Bench Litelower is betterKimi K2.7 Code by 5.340.435.1+5.3
- livebench_languageKimi K2.7 Code by 1.776.277.9+1.7
Questions people ask
Which is better, GLM-5.2 or Kimi K2.7 Code?
GLM-5.2 wins five of the seven areas where both have results: coding, agents, reasoning, math and following instructions. Kimi K2.7 Code wins facts and long documents, and costs 15% less.
Which is better for coding?
GLM-5.2. It wins 6 of the 6 coding tests both models report; Kimi K2.7 Code wins none.
Which is cheaper?
GLM-5.2 costs $1.40 per million input tokens and $4.40 per million output tokens; Kimi K2.7 Code costs $0.95 and $4.00. That makes Kimi K2.7 Code about 15% cheaper for the same work.
How do you compare the two?
We use the 34 benchmark tests both models have published scores on. The verdict counts the 24 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.