Skip to main content
BenchLM
Advertise

Model comparison

GLM-5.2 vs Kimi K2.5

Data verified

Head-to-head evidence from 29 shared benchmark results across 7 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

63.96/100
Margin
4.3pts
← winning
Kimi K2.5
Moonshot AI
59.66/100
4 category wins0 category wins

Public leaderboard positions: GLM-5.2 #37 (Estimated); Kimi K2.5 #54 (Supported). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. GLM-5.2 and Kimi K2.5 share 29 comparable benchmark results. 4 of 8 categories are comparable. 14 results are unique to GLM-5.2; 34 to Kimi K2.5.

Updated July 23, 2026
Shared results
29
GLM-5.2 only
14
Kimi K2.5 only
34
Comparable categories
4 / 8

Pick GLM-5.2 if you want the stronger benchmark profile. Kimi K2.5 only becomes the better choice if you want the cheaper token bill or you would rather avoid the extra latency and token burn of a reasoning model.

Confidence note. This is a partial-evidence comparison with 29 shared benchmark results across 7 evidence categories; 4 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.

Why this result

GLM-5.2 is clearly ahead on the BenchAlign aggregate, 63.96 to 59.66. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

GLM-5.2's sharpest advantage is in mathematics, where it averages 95.9 against 60.6. The single biggest benchmark swing on the page is Terminal-Bench 2.0, 81% to 50.8%.

GLM-5.2 is also the more expensive model on tokens at $1.40 input / $4.40 output per 1M tokens, versus $0.60 input / $3.00 output per 1M tokens for Kimi K2.5. GLM-5.2 is the reasoning model in the pair, while Kimi K2.5 is not. That usually helps on harder chain-of-thought-heavy tests, but it can also mean more latency and more token spend in real use. GLM-5.2 gives you the larger context window at 1M, compared with 256K for Kimi K2.5.

Category breakdown

Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.

Category scores and score margins for GLM-5.2 and Kimi K2.5
CategoryGLM-5.2ΔKimi K2.5
MathGLM-5.295.9Margin← 35.3Kimi K2.560.6
AgenticGLM-5.281.0Margin← 26.0Kimi K2.555.0
CodingGLM-5.262.1Margin← 2.7Kimi K2.559.4
KnowledgeGLM-5.259.6Margin← 2.7Kimi K2.556.9
ReasoningGLM-5.2Not measuredMarginNo overlapKimi K2.561.0
MultilingualGLM-5.2Not measuredMarginNo overlapKimi K2.582.3
MultimodalGLM-5.2Not measuredMarginNo overlapKimi K2.578.5
Inst. FollowingGLM-5.2Not measuredMarginNo overlapKimi K2.593.9

Decisive benchmark drivers

The largest measured benchmark gaps in this matchup, with exact reported values.

More
A · GLM-5.2B · Kimi K2.5

  1. Terminal-Bench 2.0

    Agentic
    Source ↗
    A 81%B 50.8%
    Winner: GLM-5.2Δ 30.2
    Terminal-Bench 2.0: GLM-5.2 scored 81%; Kimi K2.5 scored 50.8%. GLM-5.2 wins this benchmark.

  2. HLE

    Knowledge
    Source ↗
    A 54.7%B 30.1%
    Winner: GLM-5.2Δ 24.6
    HLE: GLM-5.2 scored 54.7%; Kimi K2.5 scored 30.1%. GLM-5.2 wins this benchmark.

  3. SWE-bench Pro

    Coding
    Source ↗
    A 62.1%B 50.7%
    Winner: GLM-5.2Δ 11.4
    SWE-bench Pro: GLM-5.2 scored 62.1%; Kimi K2.5 scored 50.7%. GLM-5.2 wins this benchmark.

  4. HMMT Feb 2026

    Math
    Source ↗
    A 92.5%B 87.1%
    Winner: GLM-5.2Δ 5.4
    HMMT Feb 2026: GLM-5.2 scored 92.5%; Kimi K2.5 scored 87.1%. GLM-5.2 wins this benchmark.

  5. GPQA

    Knowledge
    Source ↗
    A 91.2%B 87.6%
    Winner: GLM-5.2Δ 3.6
    GPQA: GLM-5.2 scored 91.2%; Kimi K2.5 scored 87.6%. GLM-5.2 wins this benchmark.

Operational comparison

Runtime and commercial metrics are compared only when both models have a complete sourced value.

MetricGLM-5.2Kimi K2.5Comparison
Input / output priceUSD per 1M tokensGLM-5.2$1.4 input / $4.4 outputKimi K2.5$0.6 input / $3 outputKimi K2.5 has the lower combined listed price.
Generation speedtokens per secondGLM-5.2Not availableKimi K2.545 tok/sA complete speed comparison is not available.
First-answer latencyseconds to first tokenGLM-5.2Not availableKimi K2.52.38 sA complete latency comparison is not available.
Context windowmaximum listed tokensGLM-5.21MKimi K2.5256KGLM-5.2 lists the larger context window.

Benchmark Deep Dive

AgenticGLM-5.2 wins
BenchmarkGLM-5.2Kimi K2.5Result
Terminal-Bench 2.0Source 81%50.8%GLM-5.2 leads
MCP AtlasSource 76.8%29.5%GLM-5.2 leads
ToolathlonSource 48.2%27.8%GLM-5.2 leads
AA Agentic IndexSource 43.1%21.7%GLM-5.2 leads
τ²-bench resultsSource 99.1%95.9%GLM-5.2 leads
GDPval-AASource 50.7%25.4%GLM-5.2 leads
GDPval-AASource 15141009GLM-5.2 leads
APEX-Agents-AASource 33.7%11.5%GLM-5.2 leads
ResearchClawBenchSource 20.7%14.0%GLM-5.2 leads
AA BriefcaseSource 1260—Not comparable
AA AutomationBenchSource 27.8%—Not comparable
AA EnterpriseOps-GymSource 42.7%—Not comparable
AA Harvey LABSource 91.0%—Not comparable
AA ITBenchSource 42.7%—Not comparable
AA Tau3 BankingSource 26.8%—Not comparable
terminalBenchHardSource 50.8%—Not comparable
aaTerminalBench21Source 77.9%—Not comparable
BrowseCompSource —60.6%Not comparable
Claw-EvalSource —52.3%Not comparable
QwenClawBenchSource —54.3%Not comparable
τ³-bench resultsSource —65.7%Not comparable
DeepSearchQASource —77.1%Not comparable
DeepPlanningSource —14.4%Not comparable
MCP-TasksSource —59.1%Not comparable
WideResearchSource —72.7%Not comparable
Gert LabsSource —45.88%Not comparable
JobBenchSource —8.7%Not comparable
CodingGLM-5.2 wins
BenchmarkGLM-5.2Kimi K2.5Result
SWE-bench ProSource 62.1%50.7%GLM-5.2 leads
NL2RepoSource 48.9%—Not comparable
Terminal-Bench 2.0Source 81.0%—Not comparable
ProgramBenchSource 63.7%—Not comparable
cursorBench32Source 55.0%—Not comparable
AA Coding IndexSource 68.8%46.8%GLM-5.2 leads
AA-SciCodeSource 50.5%49.0%GLM-5.2 leads
SWE-bench VerifiedSource —76.8%Not comparable
SWE-bench Verified*Source —70.8%Not comparable
LiveCodeBench v6Source —85.0%Not comparable
SWE MultilingualSource —73%Not comparable
SWE-RebenchSource —58.5%Not comparable
React Native EvalsSource —77.2%Not comparable
SciCodeSource —48.7%Not comparable
Reasoning
BenchmarkGLM-5.2Kimi K2.5Result
CritPtSource 20.9%3.1%GLM-5.2 leads
AA-LCRSource 71.3%65.3%GLM-5.2 leads
LongBench v2Source —61%Not comparable
KnowledgeGLM-5.2 wins
BenchmarkGLM-5.2Kimi K2.5Result
GPQASource 91.2%87.6%GLM-5.2 leads
GPQA-DSource 91.2%87.6%GLM-5.2 leads
HLESource 54.7%30.1%GLM-5.2 leads
HLE w/o toolsSource 40.5%—Not comparable
Artificial Analysis Intelligence IndexSource 51.1%35.4%GLM-5.2 leads
AA-GPQA DiamondSource 89.5%87.9%GLM-5.2 leads
AA-HLESource 40.1%29.4%GLM-5.2 leads
AA-Omniscience IndexSource 4.0%-8.1%GLM-5.2 leads
AA-Omniscience AccuracySource 25.1%34.3%Kimi K2.5 leads
AA-Omniscience Hallucination RateSource 28.1%64.6%GLM-5.2 leads
AA Openness IndexSource 44.4%—Not comparable
SuperGPQASource —69.2%Not comparable
MMLU-ProSource —87.1%Not comparable
MMLU-Pro (Arcee)Source —87.1%Not comparable
MathGLM-5.2 wins
BenchmarkGLM-5.2Kimi K2.5Result
AIME26Source 99.2%95.8%GLM-5.2 leads
HMMT Nov 2025Source 94.4%91.1%GLM-5.2 leads
HMMT Feb 2026Source 92.5%87.1%GLM-5.2 leads
MMAnswerBenchSource 91.0%81.8%GLM-5.2 leads
AIME 2025Source —96.1%Not comparable
AIME25 (Arcee)Source —96.3%Not comparable
HMMT Feb 2025Source —95.4%Not comparable
FrontierMath v2 (Tiers 1-3)Source —27.900%Not comparable
FrontierMath v2 (Tier 4)Source —4.200%Not comparable
Multilingual
BenchmarkGLM-5.2Kimi K2.5Result
MMLU-ProXSource —82.3%Not comparable
NOVA-63Source —56.0%Not comparable
Multimodal
BenchmarkGLM-5.2Kimi K2.5Result
Design Arena WebsiteSource 13401279GLM-5.2 leads
MMMU-ProSource —78.5%Not comparable
Video-MMESource —87.4%Not comparable
MMVUSource —80.4%Not comparable
VideoMMMUSource —86.6%Not comparable
AA-MMMU-ProSource —75.4%Not comparable
Inst. Following
BenchmarkGLM-5.2Kimi K2.5Result
AA-IFBenchSource 73.3%70.2%GLM-5.2 leads
IFEvalSource —93.9%Not comparable
Frequently Asked Questions (5)

Which is better, GLM-5.2 or Kimi K2.5?

GLM-5.2 is ahead on BenchLM's BenchAlign leaderboard, 63.96 to 59.66. The biggest single separator in this matchup is Terminal-Bench 2.0, where the scores are 81% and 50.8%.

Which is better for knowledge tasks, GLM-5.2 or Kimi K2.5?

GLM-5.2 has the edge for knowledge tasks in this comparison, averaging 59.6 versus 56.9. Inside this category, AA-Omniscience Hallucination Rate is the benchmark that creates the most daylight between them.

Which is better for coding, GLM-5.2 or Kimi K2.5?

GLM-5.2 has the edge for coding in this comparison, averaging 62.1 versus 59.4. Inside this category, AA Coding Index is the benchmark that creates the most daylight between them.

Which is better for math, GLM-5.2 or Kimi K2.5?

GLM-5.2 has the edge for math in this comparison, averaging 95.9 versus 60.6. Inside this category, MMAnswerBench is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, GLM-5.2 or Kimi K2.5?

GLM-5.2 has the edge for agentic tasks in this comparison, averaging 81 versus 55. Inside this category, GDPval-AA is the benchmark that creates the most daylight between them.

Self-host vs API cost

Estimates at 50,000 req/day · 1000 tokens/req average.

GLM-5.2
API / mo$4,350
Self-host / moNot listed
Break-even—
Proprietary model — self-hosting not applicable.
Kimi K2.5
API / mo$2,700
Self-host / mo$5,221
Break-even132M/day
Model the full break-even

Related Comparisons

Explore More

Z.AI Compare Pricing Methodology Find Your Best LLM Overall Rankings
Last updated: July 23, 2026

Choose a model with this week’s evidence

Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.

One email each week. Unsubscribe anytime.