Skip to main content
BenchLM
Advertise

Model comparison

Claude Opus 4.7 (Adaptive) vs Kimi K2.5

Data verified

Head-to-head evidence from 26 shared benchmark results across 6 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

66.27/100
Margin
6.6pts
← winning
Kimi K2.5
Moonshot AI
59.66/100
4 category wins1 category wins

Public leaderboard positions: Claude Opus 4.7 (Adaptive) #27 (Estimated); Kimi K2.5 #54 (Supported). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. Claude Opus 4.7 (Adaptive) and Kimi K2.5 share 26 comparable benchmark results. 5 of 8 categories are comparable. 12 results are unique to Claude Opus 4.7 (Adaptive); 37 to Kimi K2.5.

Updated July 23, 2026
Shared results
26
Claude Opus 4.7 (Adaptive) only
12
Kimi K2.5 only
37
Comparable categories
5 / 8

Pick Claude Opus 4.7 (Adaptive) if you want the stronger benchmark profile. Kimi K2.5 only becomes the better choice if multimodal & grounded is the priority or you want the cheaper token bill.

Confidence note. This is a partial-evidence comparison with 26 shared benchmark results across 6 evidence categories; 5 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.

Why this result

Claude Opus 4.7 (Adaptive) is clearly ahead on the BenchAlign aggregate, 66.27 to 59.66. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

Claude Opus 4.7 (Adaptive)'s sharpest advantage is in agentic, where it averages 75.1 against 55. The single biggest benchmark swing on the page is HLE, 54.7% to 30.1%. Kimi K2.5 does hit back in multimodal & grounded, so the answer changes if that is the part of the workload you care about most.

Claude Opus 4.7 (Adaptive) is also the more expensive model on tokens at $5.00 input / $25.00 output per 1M tokens, versus $0.60 input / $3.00 output per 1M tokens for Kimi K2.5. That is roughly 8.3x on output cost alone. Claude Opus 4.7 (Adaptive) 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. Claude Opus 4.7 (Adaptive) 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 Claude Opus 4.7 (Adaptive) and Kimi K2.5
CategoryClaude Opus 4.7 (Adaptive)ΔKimi K2.5
AgenticClaude Opus 4.7 (Adaptive)75.1Margin← 20.1Kimi K2.555.0
CodingClaude Opus 4.7 (Adaptive)78.6Margin← 19.2Kimi K2.559.4
ReasoningClaude Opus 4.7 (Adaptive)75.8Margin← 14.8Kimi K2.561.0
MultimodalClaude Opus 4.7 (Adaptive)65.1Margin→ 13.4Kimi K2.578.5
KnowledgeClaude Opus 4.7 (Adaptive)60.0Margin← 3.1Kimi K2.556.9
MathClaude Opus 4.7 (Adaptive)Not measuredMarginNo overlapKimi K2.560.6
MultilingualClaude Opus 4.7 (Adaptive)Not measuredMarginNo overlapKimi K2.582.3
Inst. FollowingClaude Opus 4.7 (Adaptive)Not measuredMarginNo overlapKimi K2.593.9

Decisive benchmark drivers

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

More
A · Claude Opus 4.7 (Adaptive)B · Kimi K2.5

  1. HLE

    Knowledge
    Source ↗
    A 54.7%B 30.1%
    Winner: Claude Opus 4.7 (Adaptive)Δ 24.6
    HLE: Claude Opus 4.7 (Adaptive) scored 54.7%; Kimi K2.5 scored 30.1%. Claude Opus 4.7 (Adaptive) wins this benchmark.

  2. BrowseComp

    Agentic
    Source ↗
    A 79.3%B 60.6%
    Winner: Claude Opus 4.7 (Adaptive)Δ 18.7
    BrowseComp: Claude Opus 4.7 (Adaptive) scored 79.3%; Kimi K2.5 scored 60.6%. Claude Opus 4.7 (Adaptive) wins this benchmark.

  3. Terminal-Bench 2.0

    Agentic
    Source ↗
    A 69.4%B 50.8%
    Winner: Claude Opus 4.7 (Adaptive)Δ 18.6
    Terminal-Bench 2.0: Claude Opus 4.7 (Adaptive) scored 69.4%; Kimi K2.5 scored 50.8%. Claude Opus 4.7 (Adaptive) wins this benchmark.

  4. SWE-bench Pro

    Coding
    Source ↗
    A 64.3%B 50.7%
    Winner: Claude Opus 4.7 (Adaptive)Δ 13.6
    SWE-bench Pro: Claude Opus 4.7 (Adaptive) scored 64.3%; Kimi K2.5 scored 50.7%. Claude Opus 4.7 (Adaptive) wins this benchmark.

  5. SWE-bench Verified

    Coding
    Source ↗
    A 87.6%B 76.8%
    Winner: Claude Opus 4.7 (Adaptive)Δ 10.8
    SWE-bench Verified: Claude Opus 4.7 (Adaptive) scored 87.6%; Kimi K2.5 scored 76.8%. Claude Opus 4.7 (Adaptive) wins this benchmark.

Operational comparison

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

MetricClaude Opus 4.7 (Adaptive)Kimi K2.5Comparison
Input / output priceUSD per 1M tokensClaude Opus 4.7 (Adaptive)$5 input / $25 outputKimi K2.5$0.6 input / $3 outputKimi K2.5 has the lower combined listed price.
Generation speedtokens per secondClaude Opus 4.7 (Adaptive)Not availableKimi K2.545 tok/sA complete speed comparison is not available.
First-answer latencyseconds to first tokenClaude Opus 4.7 (Adaptive)Not availableKimi K2.52.38 sA complete latency comparison is not available.
Context windowmaximum listed tokensClaude Opus 4.7 (Adaptive)1MKimi K2.5256KClaude Opus 4.7 (Adaptive) lists the larger context window.

Benchmark Deep Dive

AgenticClaude Opus 4.7 (Adaptive) wins
BenchmarkClaude Opus 4.7 (Adaptive)Kimi K2.5Result
Terminal-Bench 2.0Source 69.4%50.8%Claude Opus 4.7 (Adaptive) leads
BrowseCompSource 79.3%60.6%Claude Opus 4.7 (Adaptive) leads
MCP AtlasSource 77.3%29.5%Claude Opus 4.7 (Adaptive) leads
OSWorld-VerifiedSource 78%—Not comparable
CyberGymSource 73.1%—Not comparable
AA Agentic IndexSource 44.4%21.7%Claude Opus 4.7 (Adaptive) leads
τ²-bench resultsSource 88.6%95.9%Kimi K2.5 leads
GDPval-AASource 49.8%25.4%Claude Opus 4.7 (Adaptive) leads
GDPval-AASource 14951009Claude Opus 4.7 (Adaptive) leads
OSWorld 2.0Source 18.2%—Not comparable
JobBenchSource 45.9%8.7%Claude Opus 4.7 (Adaptive) leads
AA ITBenchSource 46.7%—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
ToolathlonSource —27.8%Not comparable
MCP-TasksSource —59.1%Not comparable
WideResearchSource —72.7%Not comparable
APEX-Agents-AASource —11.5%Not comparable
Gert LabsSource —45.88%Not comparable
ResearchClawBenchSource —14.0%Not comparable
CodingClaude Opus 4.7 (Adaptive) wins
BenchmarkClaude Opus 4.7 (Adaptive)Kimi K2.5Result
SWE-bench VerifiedSource 87.6%76.8%Claude Opus 4.7 (Adaptive) leads
SWE-bench ProSource 64.3%50.7%Claude Opus 4.7 (Adaptive) leads
Terminal-Bench 2.0Source 69.4%—Not comparable
AA Coding IndexSource 73.6%46.8%Claude Opus 4.7 (Adaptive) leads
AA-SciCodeSource 54.5%49.0%Claude Opus 4.7 (Adaptive) leads
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
ReasoningClaude Opus 4.7 (Adaptive) wins
BenchmarkClaude Opus 4.7 (Adaptive)Kimi K2.5Result
MRCR v2 128K-256KSource 59.2%—Not comparable
ARC-AGI-2Source 75.8%—Not comparable
AA-LCRSource 70.3%65.3%Claude Opus 4.7 (Adaptive) leads
CritPtSource 12.0%3.1%Claude Opus 4.7 (Adaptive) leads
LongBench v2Source —61%Not comparable
KnowledgeClaude Opus 4.7 (Adaptive) wins
BenchmarkClaude Opus 4.7 (Adaptive)Kimi K2.5Result
GPQASource 94.2%87.6%Claude Opus 4.7 (Adaptive) leads
GPQA-DSource 94.2%87.6%Claude Opus 4.7 (Adaptive) leads
HLESource 54.7%30.1%Claude Opus 4.7 (Adaptive) leads
HLE w/o toolsSource 46.9%—Not comparable
Artificial Analysis Intelligence IndexSource 53.5%35.4%Claude Opus 4.7 (Adaptive) leads
AA-GPQA DiamondSource 91.4%87.9%Claude Opus 4.7 (Adaptive) leads
AA-HLESource 39.6%29.4%Claude Opus 4.7 (Adaptive) leads
AA-Omniscience IndexSource 26.2%-8.1%Claude Opus 4.7 (Adaptive) leads
AA-Omniscience AccuracySource 45.8%34.3%Claude Opus 4.7 (Adaptive) leads
AA-Omniscience Hallucination RateSource 36.2%64.6%Claude Opus 4.7 (Adaptive) leads
SuperGPQASource —69.2%Not comparable
MMLU-ProSource —87.1%Not comparable
MMLU-Pro (Arcee)Source —87.1%Not comparable
Math
BenchmarkClaude Opus 4.7 (Adaptive)Kimi K2.5Result
FrontierMath (legacy)Source 43.8%—Not comparable
AIME 2025Source —96.1%Not comparable
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
FrontierMath v2 (Tiers 1-3)Source —27.900%Not comparable
FrontierMath v2 (Tier 4)Source —4.200%Not comparable
Multilingual
BenchmarkClaude Opus 4.7 (Adaptive)Kimi K2.5Result
MMLU-ProXSource —82.3%Not comparable
NOVA-63Source —56.0%Not comparable
MultimodalKimi K2.5 wins
BenchmarkClaude Opus 4.7 (Adaptive)Kimi K2.5Result
OfficeQA ProSource 43.6%—Not comparable
CharXivSource 91%—Not comparable
CharXiv w/o toolsSource 82.1%—Not comparable
AA-MMMU-ProSource 78.8%75.4%Claude Opus 4.7 (Adaptive) leads
Design Arena WebsiteSource 13251279Claude Opus 4.7 (Adaptive) leads
MMMU-ProSource —78.5%Not comparable
Video-MMESource —87.4%Not comparable
MMVUSource —80.4%Not comparable
VideoMMMUSource —86.6%Not comparable
Inst. Following
BenchmarkClaude Opus 4.7 (Adaptive)Kimi K2.5Result
AA-IFBenchSource 58.6%70.2%Kimi K2.5 leads
IFEvalSource —93.9%Not comparable
Frequently Asked Questions (6)

Which is better, Claude Opus 4.7 (Adaptive) or Kimi K2.5?

Claude Opus 4.7 (Adaptive) is ahead on BenchLM's BenchAlign leaderboard, 66.27 to 59.66. The biggest single separator in this matchup is HLE, where the scores are 54.7% and 30.1%.

Which is better for knowledge tasks, Claude Opus 4.7 (Adaptive) or Kimi K2.5?

Claude Opus 4.7 (Adaptive) has the edge for knowledge tasks in this comparison, averaging 60 versus 56.9. Inside this category, AA-Omniscience Index is the benchmark that creates the most daylight between them.

Which is better for coding, Claude Opus 4.7 (Adaptive) or Kimi K2.5?

Claude Opus 4.7 (Adaptive) has the edge for coding in this comparison, averaging 78.6 versus 59.4. Inside this category, AA Coding Index is the benchmark that creates the most daylight between them.

Which is better for reasoning, Claude Opus 4.7 (Adaptive) or Kimi K2.5?

Claude Opus 4.7 (Adaptive) has the edge for reasoning in this comparison, averaging 75.8 versus 61. Inside this category, CritPt is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, Claude Opus 4.7 (Adaptive) or Kimi K2.5?

Claude Opus 4.7 (Adaptive) has the edge for agentic tasks in this comparison, averaging 75.1 versus 55. Inside this category, GDPval-AA is the benchmark that creates the most daylight between them.

Which is better for multimodal and grounded tasks, Claude Opus 4.7 (Adaptive) or Kimi K2.5?

Kimi K2.5 has the edge for multimodal and grounded tasks in this comparison, averaging 78.5 versus 65.1. Inside this category, Design Arena Website 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.

Claude Opus 4.7 (Adaptive)
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

Anthropic 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.