Skip to main content
BenchLM
Advertise

Model comparison

GLM-4.7 vs Kimi K2.5

Data verified

Head-to-head evidence from 27 shared benchmark results across 7 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

61.16/100
Margin
1.5pts
← winning
Kimi K2.5
Moonshot AI
59.66/100
1 category wins3 category wins

Public leaderboard positions: GLM-4.7 #42 (Supported); Kimi K2.5 #54 (Supported). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. GLM-4.7 and Kimi K2.5 share 27 comparable benchmark results. 4 of 8 categories are comparable. 3 results are unique to GLM-4.7; 36 to Kimi K2.5.

Updated July 20, 2026
Shared results
27
GLM-4.7 only
3
Kimi K2.5 only
36
Comparable categories
4 / 8

Pick GLM-4.7 if you want the stronger benchmark profile. Kimi K2.5 only becomes the better choice if mathematics is the priority or you need the larger 256K context window.

Confidence note. This is a partial-evidence comparison with 27 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-4.7 has the cleaner BenchAlign overall profile here, landing at 61.16 versus 59.66. It is a real lead, but still close enough that category-level strengths matter more than the headline number.

GLM-4.7's sharpest advantage is in coding, where it averages 75.4 against 59.4. The single biggest benchmark swing on the page is FrontierMath v2 (Tiers 1-3), 2.439% to 27.900%. Kimi K2.5 does hit back in mathematics, so the answer changes if that is the part of the workload you care about most.

Kimi K2.5 is also the more expensive model on tokens at $0.60 input / $3.00 output per 1M tokens, versus $0.00 input / $0.00 output per 1M tokens for GLM-4.7. That is roughly Infinityx on output cost alone. GLM-4.7 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. Kimi K2.5 gives you the larger context window at 256K, compared with 200K for GLM-4.7.

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-4.7 and Kimi K2.5
CategoryGLM-4.7ΔKimi K2.5
MathGLM-4.71.8Margin→ 58.8Kimi K2.560.6
CodingGLM-4.775.4Margin← 16.0Kimi K2.559.4
AgenticGLM-4.745.7Margin→ 9.3Kimi K2.555.0
KnowledgeGLM-4.751.8Margin→ 5.1Kimi K2.556.9
ReasoningGLM-4.7Not measuredMarginNo overlapKimi K2.561.0
MultilingualGLM-4.7Not measuredMarginNo overlapKimi K2.582.3
MultimodalGLM-4.7Not measuredMarginNo overlapKimi K2.578.5
Inst. FollowingGLM-4.7Not measuredMarginNo overlapKimi K2.593.9

Decisive benchmark drivers

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

More
A · GLM-4.7B · Kimi K2.5

  1. FrontierMath v2 (Tiers 1-3)

    Math
    Source ↗
    A 2.439%B 27.900%
    Winner: Kimi K2.5Δ 25.5
    FrontierMath v2 (Tiers 1-3): GLM-4.7 scored 2.439%; Kimi K2.5 scored 27.900%. Kimi K2.5 wins this benchmark.

  2. Terminal-Bench 2.0

    Agentic
    Source ↗
    A 41%B 50.8%
    Winner: Kimi K2.5Δ 9.8
    Terminal-Bench 2.0: GLM-4.7 scored 41%; Kimi K2.5 scored 50.8%. Kimi K2.5 wins this benchmark.

  3. BrowseComp

    Agentic
    Source ↗
    A 52%B 60.6%
    Winner: Kimi K2.5Δ 8.6
    BrowseComp: GLM-4.7 scored 52%; Kimi K2.5 scored 60.6%. Kimi K2.5 wins this benchmark.

  4. HLE

    Knowledge
    Source ↗
    A 24.8%B 30.1%
    Winner: Kimi K2.5Δ 5.3
    HLE: GLM-4.7 scored 24.8%; Kimi K2.5 scored 30.1%. Kimi K2.5 wins this benchmark.

  5. FrontierMath v2 (Tier 4)

    Math
    Source ↗
    A 0.000%B 4.200%
    Winner: Kimi K2.5Δ 4.2
    FrontierMath v2 (Tier 4): GLM-4.7 scored 0.000%; Kimi K2.5 scored 4.200%. Kimi K2.5 wins this benchmark.

Operational comparison

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

MetricGLM-4.7Kimi K2.5Comparison
Input / output priceUSD per 1M tokensGLM-4.7$0 input / $0 outputKimi K2.5$0.6 input / $3 outputGLM-4.7 has the lower combined listed price.
Generation speedtokens per secondGLM-4.782 tok/sKimi K2.545 tok/sGLM-4.7 has the higher measured throughput.
First-answer latencyseconds to first tokenGLM-4.71.10 sKimi K2.52.38 sGLM-4.7 reaches the first token sooner.
Context windowmaximum listed tokensGLM-4.7200KKimi K2.5256KKimi K2.5 lists the larger context window.

Benchmark Deep Dive

AgenticKimi K2.5 wins
BenchmarkGLM-4.7Kimi K2.5Result
Terminal-Bench 2.0Source 41%50.8%Kimi K2.5 leads
BrowseCompSource 52%60.6%Kimi K2.5 leads
VITA-BenchSource 15.5%—Not comparable
AA Agentic IndexSource 25.4%21.7%GLM-4.7 leads
τ²-bench resultsSource 95.9%95.9%Tie
Gert LabsSource 39.95%45.88%Kimi K2.5 leads
GDPval-AASource 33.3%25.4%GLM-4.7 leads
GDPval-AASource 11651009GLM-4.7 leads
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
ToolathlonSource —27.8%Not comparable
MCP AtlasSource —29.5%Not comparable
MCP-TasksSource —59.1%Not comparable
WideResearchSource —72.7%Not comparable
APEX-Agents-AASource —11.5%Not comparable
ResearchClawBenchSource —14.0%Not comparable
JobBenchSource —8.7%Not comparable
CodingGLM-4.7 wins
BenchmarkGLM-4.7Kimi K2.5Result
SWE-bench VerifiedSource 73.8%76.8%Kimi K2.5 leads
LiveCodeBenchSource 84.9%—Not comparable
SWE-RebenchSource 58.7%58.5%GLM-4.7 leads
AA Coding IndexSource 45.3%46.8%Kimi K2.5 leads
AA-SciCodeSource 45.1%49.0%Kimi K2.5 leads
AA LiveCodeBenchSource 89.4%—Not comparable
SWE-bench Verified*Source —70.8%Not comparable
LiveCodeBench v6Source —85.0%Not comparable
SWE-bench ProSource —50.7%Not comparable
SWE MultilingualSource —73%Not comparable
React Native EvalsSource —77.2%Not comparable
SciCodeSource —48.7%Not comparable
Reasoning
BenchmarkGLM-4.7Kimi K2.5Result
AA-LCRSource 64.0%65.3%Kimi K2.5 leads
CritPtSource 1.7%3.1%Kimi K2.5 leads
LongBench v2Source —61%Not comparable
KnowledgeKimi K2.5 wins
BenchmarkGLM-4.7Kimi K2.5Result
GPQASource 85.7%87.6%Kimi K2.5 leads
MMLU-ProSource 84.3%87.1%Kimi K2.5 leads
HLESource 24.8%30.1%Kimi K2.5 leads
Artificial Analysis Intelligence IndexSource 33.7%35.4%Kimi K2.5 leads
AA-GPQA DiamondSource 85.9%87.9%Kimi K2.5 leads
AA-HLESource 25.1%29.4%Kimi K2.5 leads
AA-Omniscience IndexSource -34.6%-8.1%Kimi K2.5 leads
AA-Omniscience AccuracySource 29.3%34.3%Kimi K2.5 leads
AA-Omniscience Hallucination RateSource 90.3%64.6%Kimi K2.5 leads
GPQA-DSource —87.6%Not comparable
SuperGPQASource —69.2%Not comparable
MMLU-Pro (Arcee)Source —87.1%Not comparable
MathKimi K2.5 wins
BenchmarkGLM-4.7Kimi K2.5Result
AIME 2025Source 95.7%96.1%Kimi K2.5 leads
FrontierMath v2 (Tiers 1-3)Source 2.439%27.900%Kimi K2.5 leads
FrontierMath v2 (Tier 4)Source 0.000%4.200%Kimi K2.5 leads
AIME26Source —95.8%Not comparable
AIME25 (Arcee)Source —96.3%Not comparable
HMMT Feb 2025Source —95.4%Not comparable
HMMT Nov 2025Source —91.1%Not comparable
HMMT Feb 2026Source —87.1%Not comparable
MMAnswerBenchSource —81.8%Not comparable
Multilingual
BenchmarkGLM-4.7Kimi K2.5Result
MMLU-ProXSource —82.3%Not comparable
NOVA-63Source —56.0%Not comparable
Multimodal
BenchmarkGLM-4.7Kimi K2.5Result
Design Arena WebsiteSource 12581282Kimi K2.5 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-4.7Kimi K2.5Result
AA-IFBenchSource 67.9%70.2%Kimi K2.5 leads
IFEvalSource —93.9%Not comparable
Frequently Asked Questions (5)

Which is better, GLM-4.7 or Kimi K2.5?

GLM-4.7 is ahead on BenchLM's BenchAlign leaderboard, 61.16 to 59.66. The biggest single separator in this matchup is FrontierMath v2 (Tiers 1-3), where the scores are 2.439% and 27.900%.

Which is better for knowledge tasks, GLM-4.7 or Kimi K2.5?

Kimi K2.5 has the edge for knowledge tasks in this comparison, averaging 56.9 versus 51.8. Inside this category, AA-Omniscience Index is the benchmark that creates the most daylight between them.

Which is better for coding, GLM-4.7 or Kimi K2.5?

GLM-4.7 has the edge for coding in this comparison, averaging 75.4 versus 59.4. Inside this category, AA-SciCode is the benchmark that creates the most daylight between them.

Which is better for math, GLM-4.7 or Kimi K2.5?

Kimi K2.5 has the edge for math in this comparison, averaging 60.6 versus 1.8. Inside this category, FrontierMath v2 (Tiers 1-3) is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, GLM-4.7 or Kimi K2.5?

Kimi K2.5 has the edge for agentic tasks in this comparison, averaging 55 versus 45.7. 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-4.7
API / mo$0
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 20, 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.