Skip to main content

Model comparison

Kimi K2.5 (Reasoning) vs Qwen3.6-27B

Data verified

Head-to-head evidence from 22 shared benchmark results across 6 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

59.35/100
Margin
5.5pts
← winning
53.82/100
2 category wins2 category wins

Public leaderboard positions: Kimi K2.5 (Reasoning) #57 (Estimated); Qwen3.6-27B #93 (Estimated). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. Kimi K2.5 (Reasoning) and Qwen3.6-27B share 22 comparable benchmark results. 4 of 8 categories are comparable. 5 results are unique to Kimi K2.5 (Reasoning); 32 to Qwen3.6-27B.

Updated July 21, 2026
Shared results
22
Kimi K2.5 (Reasoning) only
5
Qwen3.6-27B only
32
Comparable categories
4 / 8

Pick Kimi K2.5 (Reasoning) if you want the stronger benchmark profile. Qwen3.6-27B only becomes the better choice if agentic is the priority or you want the cheaper token bill.

Confidence note. This is a partial-evidence comparison with 22 shared benchmark results across 6 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

Kimi K2.5 (Reasoning) is clearly ahead on the BenchAlign aggregate, 59.35 to 53.82. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

Kimi K2.5 (Reasoning)'s sharpest advantage is in knowledge, where it averages 87.2 against 53.3. The single biggest benchmark swing on the page is Terminal-Bench 2.0, 50.8% to 59.3%. Qwen3.6-27B does hit back in agentic, so the answer changes if that is the part of the workload you care about most.

Kimi K2.5 (Reasoning) 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 Qwen3.6-27B. That is roughly Infinityx on output cost alone. Qwen3.6-27B gives you the larger context window at 262K, compared with 128K for Kimi K2.5 (Reasoning).

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 Kimi K2.5 (Reasoning) and Qwen3.6-27B
CategoryKimi K2.5 (Reasoning)ΔQwen3.6-27B
KnowledgeKimi K2.5 (Reasoning)87.2Margin 33.9Qwen3.6-27B53.3
AgenticKimi K2.5 (Reasoning)55.0Margin 4.3Qwen3.6-27B59.3
MultimodalKimi K2.5 (Reasoning)78.5Margin 1.8Qwen3.6-27B76.7
CodingKimi K2.5 (Reasoning)76.8Margin 0.7Qwen3.6-27B77.5
MathKimi K2.5 (Reasoning)Not measuredMarginNo overlapQwen3.6-27B89.2

Decisive benchmark drivers

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

More
A · Kimi K2.5 (Reasoning)B · Qwen3.6-27B
  1. Terminal-Bench 2.0

    Agentic
    Source ↗
    A 50.8%B 59.3%
    Winner: Qwen3.6-27BΔ 8.5
    Terminal-Bench 2.0: Kimi K2.5 (Reasoning) scored 50.8%; Qwen3.6-27B scored 59.3%. Qwen3.6-27B wins this benchmark.
  2. MMMU-Pro

    Multimodal
    Source ↗
    A 78.5%B 75.8%
    Winner: Kimi K2.5 (Reasoning)Δ 2.7
    MMMU-Pro: Kimi K2.5 (Reasoning) scored 78.5%; Qwen3.6-27B scored 75.8%. Kimi K2.5 (Reasoning) wins this benchmark.
  3. MMLU-Pro

    Knowledge
    Source ↗
    A 87.1%B 86.2%
    Winner: Kimi K2.5 (Reasoning)Δ 0.9
    MMLU-Pro: Kimi K2.5 (Reasoning) scored 87.1%; Qwen3.6-27B scored 86.2%. Kimi K2.5 (Reasoning) wins this benchmark.
  4. SWE-bench Verified

    Coding
    Source ↗
    A 76.8%B 77.2%
    Winner: Qwen3.6-27BΔ 0.4
    SWE-bench Verified: Kimi K2.5 (Reasoning) scored 76.8%; Qwen3.6-27B scored 77.2%. Qwen3.6-27B wins this benchmark.
  5. GPQA

    Knowledge
    Source ↗
    A 87.6%B 87.8%
    Winner: Qwen3.6-27BΔ 0.2
    GPQA: Kimi K2.5 (Reasoning) scored 87.6%; Qwen3.6-27B scored 87.8%. Qwen3.6-27B wins this benchmark.

Operational comparison

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

MetricKimi K2.5 (Reasoning)Qwen3.6-27BComparison
Input / output priceUSD per 1M tokensKimi K2.5 (Reasoning)$0.6 input / $3 outputQwen3.6-27B$0 input / $0 outputQwen3.6-27B has the lower combined listed price.
Generation speedtokens per secondKimi K2.5 (Reasoning)Not availableQwen3.6-27BNot availableA complete speed comparison is not available.
First-answer latencyseconds to first tokenKimi K2.5 (Reasoning)Not availableQwen3.6-27BNot availableA complete latency comparison is not available.
Context windowmaximum listed tokensKimi K2.5 (Reasoning)128KQwen3.6-27B262KQwen3.6-27B lists the larger context window.

Benchmark Deep Dive

AgenticQwen3.6-27B wins
BenchmarkKimi K2.5 (Reasoning)Qwen3.6-27BResult
Terminal-Bench 2.0Source 50.8%59.3%Qwen3.6-27B leads
BrowseCompSource 60.6%Not comparable
APEX-Agents-AASource 11.5%Not comparable
τ²-bench resultsSource 95.9%94.2%Kimi K2.5 (Reasoning) leads
Gert LabsSource 32.58%54.84%Qwen3.6-27B leads
AA Agentic IndexSource 21.7%27.0%Qwen3.6-27B leads
GDPval-AASource 25.4%32.0%Qwen3.6-27B leads
GDPval-AASource 10091140Qwen3.6-27B leads
Claw-EvalSource 72.4%Not comparable
QwenClawBenchSource 53.4%Not comparable
QwenWebBenchSource 1487Not comparable
AndroidWorldSource 70.3%Not comparable
CodingQwen3.6-27B wins
BenchmarkKimi K2.5 (Reasoning)Qwen3.6-27BResult
SWE-bench VerifiedSource 76.8%77.2%Qwen3.6-27B leads
Vibe Code BenchSource 17.54%Not comparable
AA-SciCodeSource 49.0%39.8%Kimi K2.5 (Reasoning) leads
AA Coding IndexSource 46.8%53.7%Qwen3.6-27B leads
SWE MultilingualSource 71.3%Not comparable
SWE-bench ProSource 53.5%Not comparable
Terminal-Bench 2.0Source 59.3%Not comparable
LiveCodeBenchSource 83.9%Not comparable
NL2RepoSource 36.2%Not comparable
Reasoning
BenchmarkKimi K2.5 (Reasoning)Qwen3.6-27BResult
AA-LCRSource 65.3%68.7%Qwen3.6-27B leads
CritPtSource 3.1%1.1%Kimi K2.5 (Reasoning) leads
KnowledgeKimi K2.5 (Reasoning) wins
BenchmarkKimi K2.5 (Reasoning)Qwen3.6-27BResult
GPQASource 87.6%87.8%Qwen3.6-27B leads
MMLU-ProSource 87.1%86.2%Kimi K2.5 (Reasoning) leads
Artificial Analysis Intelligence IndexSource 35.4%37.0%Qwen3.6-27B leads
AA-GPQA DiamondSource 87.9%84.2%Kimi K2.5 (Reasoning) leads
AA-HLESource 29.4%21.6%Kimi K2.5 (Reasoning) leads
AA-Omniscience IndexSource -8.1%-19.8%Kimi K2.5 (Reasoning) leads
AA-Omniscience AccuracySource 34.3%19.2%Kimi K2.5 (Reasoning) leads
AA-Omniscience Hallucination RateSource 64.6%48.3%Qwen3.6-27B leads
MMLU-ReduxSource 93.5%Not comparable
SuperGPQASource 66%Not comparable
C-EvalSource 91.4%Not comparable
HLESource 24%Not comparable
Math
BenchmarkKimi K2.5 (Reasoning)Qwen3.6-27BResult
AIME 2025Source 96.1%Not comparable
HMMT Feb 2025Source 93.8%Not comparable
HMMT Nov 2025Source 90.7%Not comparable
HMMT Feb 2026Source 84.3%Not comparable
MMAnswerBenchSource 80.8%Not comparable
AIME26Source 94.1%Not comparable
MultimodalKimi K2.5 (Reasoning) wins
BenchmarkKimi K2.5 (Reasoning)Qwen3.6-27BResult
MMMU-ProSource 78.5%75.8%Kimi K2.5 (Reasoning) leads
AA-MMMU-ProSource 75.4%74.6%Kimi K2.5 (Reasoning) leads
Design Arena WebsiteSource 1279Not comparable
MMMUSource 82.9%Not comparable
RealWorldQASource 84.1%Not comparable
DynaMathSource 85.6%Not comparable
MStarSource 81.4%Not comparable
SimpleVQASource 56.1%Not comparable
CharXivSource 78.4%Not comparable
CC-OCRSource 81.2%Not comparable
CountBenchSource 97.8%Not comparable
RefCOCO (avg)Source 92.5%Not comparable
ERQASource 62.5%Not comparable
Video-MME (with subtitle)Source 87.7%Not comparable
VideoMMMUSource 84.4%Not comparable
MLVU (M-Avg)Source 86.6%Not comparable
V*Source 94.7%Not comparable
Inst. Following
BenchmarkKimi K2.5 (Reasoning)Qwen3.6-27BResult
AA-IFBenchSource 70.2%67.6%Kimi K2.5 (Reasoning) leads
Frequently Asked Questions (5)

Which is better, Kimi K2.5 (Reasoning) or Qwen3.6-27B?

Kimi K2.5 (Reasoning) is ahead on BenchLM's BenchAlign leaderboard, 59.35 to 53.82. The biggest single separator in this matchup is Terminal-Bench 2.0, where the scores are 50.8% and 59.3%.

Which is better for knowledge tasks, Kimi K2.5 (Reasoning) or Qwen3.6-27B?

Kimi K2.5 (Reasoning) has the edge for knowledge tasks in this comparison, averaging 87.2 versus 53.3. Inside this category, AA-Omniscience Hallucination Rate is the benchmark that creates the most daylight between them.

Which is better for coding, Kimi K2.5 (Reasoning) or Qwen3.6-27B?

Qwen3.6-27B has the edge for coding in this comparison, averaging 77.5 versus 76.8. Inside this category, AA-SciCode is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, Kimi K2.5 (Reasoning) or Qwen3.6-27B?

Qwen3.6-27B has the edge for agentic tasks in this comparison, averaging 59.3 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, Kimi K2.5 (Reasoning) or Qwen3.6-27B?

Kimi K2.5 (Reasoning) has the edge for multimodal and grounded tasks in this comparison, averaging 78.5 versus 76.7. Inside this category, MMMU-Pro 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.

Kimi K2.5 (Reasoning)
API / mo$2,700
Self-host / moNot listed
Break-even
Proprietary model — self-hosting not applicable.
Qwen3.6-27B
API / mo$0
Self-host / mo$429
Break-even
Model the full break-even

Related Comparisons

Last updated: July 21, 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.