Model A

Moonshot AI

80.5/100

Supported · Public rank #5

90% interval 77.7–83.4

Kimi K3 vs Qwen3.6-27B

Updated August 19, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.

Model B

Qwen3.6-27B

Alibaba

53.7/100

Estimated · Public rank #106

90% interval 42.2–65.2

Decision reading

Kimi K3 has the higher public score, 80.53 versus 53.71, and the 90% score intervals do not overlap.

5 results are shared. Category rows based on different benchmark sets are marked directional and do not name a winner.

Which one for your work

Recommendations appear only when a shared evidence basis or an explicit operating constraint supports the call. Secondary and unsupported use cases stay disclosed below the initial list.

Long documents
Prompts that approach the documented context limit
Kimi K3
Kimi K3 has the larger documented context window.
Confidence: documented

Show secondary and unsupported calls

Coding work
Code generation, repair, and software-engineering tasks
Not enough matched evidence
No shared weighted benchmark basis supports a winner.
Confidence: limited
Agentic work
Tool use, computer use, and multi-step task completion
Not enough matched evidence
The category averages use different weighted benchmark sets, so they are directional rather than like-for-like.
Confidence: limited
Chat turn cost
1K fresh input + 500 output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates
Cache-heavy agent loop cost
200K cached + 20K fresh input + 10K output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates
Repository review cost
50K fresh input + 3K output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates

What is actually comparable

Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.

35 Kimi K3 only

5 Shared

33 Qwen3.6-27B only

Shared results: 5
Kimi K3 only: 35
Qwen3.6-27B only: 33
Like-for-like categories: 0 / 8

3 categories use different evidence sets. Those rows remain visible for coverage context but do not name a winner.

Category results, on a stated basis

Each row states whether both averages use the same weighted benchmark set. Directional and not-comparable rows remain visible, but they never receive a winner in this template.

Agentic

Directional only

Kimi K3: 89.5
Qwen3.6-27B: 59.3
Weighted basis: 2 vs 1 rows
Reading: Directional only

Knowledge

Directional only

Kimi K3: 61.0
Qwen3.6-27B: 53.3
Weighted basis: 2 vs 4 rows
Reading: Directional only

Multimodal

Directional only

Kimi K3: 78.5
Qwen3.6-27B: 76.7
Weighted basis: 3 vs 2 rows
Reading: Directional only

Coding

Not comparable

Kimi K3: Not measured
Qwen3.6-27B: 77.5
Weighted basis: 0 vs 3 rows
Reading: Not comparable

Reasoning

Not comparable

Kimi K3: Not measured
Qwen3.6-27B: Not measured
Weighted basis: 0 vs 0 rows
Reading: Not comparable

Math

Not comparable

Kimi K3: Not measured
Qwen3.6-27B: 89.2
Weighted basis: 0 vs 2 rows
Reading: Not comparable

Multilingual

Not comparable

Kimi K3: Not measured
Qwen3.6-27B: Not measured
Weighted basis: 0 vs 0 rows
Reading: Not comparable

Instruction following

Not comparable

Kimi K3: Not measured
Qwen3.6-27B: Not measured
Weighted basis: 0 vs 0 rows
Reading: Not comparable

Category averages with the server-provided evidence basis for Kimi K3 and Qwen3.6-27B
Category	Kimi K3	Qwen3.6-27B	Weighted basis	Reading
Agentic	89.5	59.3	Directional only2 vs 1 rows	Directional only
Knowledge	61.0	53.3	Directional only2 vs 4 rows	Directional only
Multimodal	78.5	76.7	Directional only3 vs 2 rows	Directional only
Coding	Not measured	77.5	Not comparable0 vs 3 rows	Not comparable
Reasoning	Not measured	Not measured	Not comparable0 vs 0 rows	Not comparable
Math	Not measured	89.2	Not comparable0 vs 2 rows	Not comparable
Multilingual	Not measured	Not measured	Not comparable0 vs 0 rows	Not comparable
Instruction following	Not measured	Not measured	Not comparable0 vs 0 rows	Not comparable

Shape of the matched evidence

Only shared public evidence is shown. Sparse evidence stays a ruled list rather than being closed into a radar shape.

Too few matched category axes support a radar. The ruled list below shows only shared benchmark results; positions use each benchmark’s normalized display scale when available.

HLE
Knowledge
Kimi K3: 56%Qwen3.6-27B: 24%Normalized gap 32.0Kimi K3 source Qwen3.6-27B source
Terminal-Bench 2.0
Agentic
Kimi K3: 88.3%Qwen3.6-27B: 59.3%Normalized gap 29.0Kimi K3 source Qwen3.6-27B source
CharXiv
Multimodal
Kimi K3: 91.3%Qwen3.6-27B: 78.4%Normalized gap 12.9Kimi K3 source Qwen3.6-27B source
MMMU-Pro
Multimodal
Kimi K3: 81.6%Qwen3.6-27B: 75.8%Normalized gap 5.8Kimi K3 source Qwen3.6-27B source
GPQA
Knowledge
Kimi K3: 93.5%Qwen3.6-27B: 87.8%Normalized gap 5.7Kimi K3 source Qwen3.6-27B source

What each workload costs

Three fixed token mixes turn per-token rates into comparable decisions. Each scenario states context fit and whether cached input had to fall back to the published list-input rate.

Chat turn

1K fresh input + 500 output tokens

Kimi K3: $0.0105; Fits in one request
Qwen3.6-27B: Self-hosted; infrastructure cost varies; Fits in one request

Qwen3.6-27B has no comparable published API token rate.

Repository review

50K fresh input + 3K output tokens

Kimi K3: $0.195; Fits in one request
Qwen3.6-27B: Self-hosted; infrastructure cost varies; Fits in one request

Qwen3.6-27B has no comparable published API token rate.

Cache-heavy agent loop

200K cached + 20K fresh input + 10K output tokens

Kimi K3: $0.27; Fits in one request
Qwen3.6-27B: Self-hosted; infrastructure cost varies; Fits in one request; Cached-input rate unavailable

Qwen3.6-27B has no comparable published API token rate.

Specification differences

Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.

SpecificationKimi K3Qwen3.6-27B

Context window

Maximum documented context; output-token limits may be lower.

Kimi K3

1.05M

Qwen3.6-27B

262K

API model ID

Kimi K3

Not sourced

Qwen3.6-27B

Not sourced

Cached-input rate

A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.

Kimi K3

$0.3 per 1M cached input tokens

Qwen3.6-27B

No comparable hosted API rate

Documented inputs

Kimi K3

Not sourced

Qwen3.6-27B

Not sourced

Documented outputs

Kimi K3

Not sourced

Qwen3.6-27B

Not sourced

Provider availability

Kimi K3

Not sourced

Qwen3.6-27B

Not sourced

Reasoning profile

Kimi K3

Reasoning

Qwen3.6-27B

Reasoning

Weight access

Kimi K3

Pending

Qwen3.6-27B

Open Weight

License

Kimi K3

Pending

Qwen3.6-27B

Open Weight

Release date

Kimi K3

2026-07-16

Qwen3.6-27B

2026-04-21

If you already use one of these models

Deployment change: The models list different providers, so authentication, endpoint behavior, limits, and feature support may change.
Quality signal: Kimi K3 has the higher public score, 80.53 versus 53.71, and the 90% score intervals do not overlap.
Workload cost: A complete comparable API-rate estimate is not available for both models.
Context tradeoff: Kimi K3 has the larger documented window (1.05M).

Run the same representative tasks against both endpoints before changing production traffic.

Self-host vs API cost

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

Kimi K3

API / mo$0

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

Benchmark evidence

The full public result ledger is available for audit without forcing a wide desktop table onto a phone.

Browse raw public benchmark evidence73 rows

Agentic

Terminal-Bench 2.0
Kimi K388.3%
Source
Qwen3.6-27B59.3%
Source
Kimi K3 leads this result
BrowseComp
Kimi K391.2%
Source
Qwen3.6-27B—
Not directly comparable
DeepSearchQA
Kimi K395.0%
Source
Qwen3.6-27B—
Not directly comparable
Toolathlon-Verified
Kimi K373.2%
Source
Qwen3.6-27B—
Not directly comparable
MCP Atlas
Kimi K384.2%
Source
Qwen3.6-27B—
Not directly comparable
AutomationBench
Kimi K330.8%
Source
Qwen3.6-27B—
Not directly comparable
JobBench
Kimi K352.9%
Source
Qwen3.6-27B—
Not directly comparable
APEX-Agents
Kimi K337.6%
Source
Qwen3.6-27B—
Not directly comparable
SpreadsheetBench 2
Kimi K334.8%
Source
Qwen3.6-27B—
Not directly comparable
DECK-Bench
Kimi K373.5%
Source
Qwen3.6-27B—
Not directly comparable
Claw-Eval
Kimi K3—
Qwen3.6-27B72.4%
Source
Not directly comparable
QwenClawBench
Kimi K3—
Qwen3.6-27B53.4%
Source
Not directly comparable
QwenWebBench
Kimi K3—
Qwen3.6-27B1487
Source
Not directly comparable
AndroidWorld
Kimi K3—
Qwen3.6-27B70.3%
Source
Not directly comparable
Gert Labs
Kimi K3—
Qwen3.6-27B54.84%
Source
Not directly comparable

Coding

deepSwe
Kimi K367.5%
Source
Qwen3.6-27B—
Not directly comparable
cursorBench32
Kimi K360.8%
Source
Qwen3.6-27B—
Not directly comparable
FrontierSWE
Kimi K381.2%
Source
Qwen3.6-27B—
Not directly comparable
ProgramBench
Kimi K377.8%
Source
Qwen3.6-27B—
Not directly comparable
Kimi Code Bench v2
Kimi K372.9%
Source
Qwen3.6-27B—
Not directly comparable
sweMarathon
Kimi K342%
Source
Qwen3.6-27B—
Not directly comparable
PostTrain Bench
Kimi K336.6%
Source
Qwen3.6-27B—
Not directly comparable
MLS-Bench Lite
Kimi K348.3%
Source
Qwen3.6-27B—
Not directly comparable
VulcanBench v3
Kimi K373.9%
Source
Qwen3.6-27B—
Not directly comparable
APEX-SWE
Kimi K348.0%
Source
Qwen3.6-27B—
Not directly comparable
EEBench
Kimi K338.3%
Source
Qwen3.6-27B—
Not directly comparable
InferenceEval
Kimi K339.8%
Source
Qwen3.6-27B—
Not directly comparable
KernelBench Internal
Kimi K368.5%
Source
Qwen3.6-27B—
Not directly comparable
SWE-bench Verified
Kimi K3—
Qwen3.6-27B77.2%
Source
Not directly comparable
SWE Multilingual
Kimi K3—
Qwen3.6-27B71.3%
Source
Not directly comparable
SWE-bench Pro
Kimi K3—
Qwen3.6-27B53.5%
Source
Not directly comparable
Terminal-Bench 2.0
Kimi K3—
Qwen3.6-27B59.3%
Source
Not directly comparable
LiveCodeBench
Kimi K3—
Qwen3.6-27B83.9%
Source
Not directly comparable
NL2Repo
Kimi K3—
Qwen3.6-27B36.2%
Source
Not directly comparable

Knowledge

GPQA
Kimi K393.5%
Source
Qwen3.6-27B87.8%
Source
Kimi K3 leads this result
GPQA-D
Kimi K393.5%
Source
Qwen3.6-27B—
Not directly comparable
HLE
Kimi K356%
Source
Qwen3.6-27B24%
Source
Kimi K3 leads this result
HLE w/o tools
Kimi K343.5%
Source
Qwen3.6-27B—
Not directly comparable
MMLU-Pro
Kimi K3—
Qwen3.6-27B86.2%
Source
Not directly comparable
MMLU-Redux
Kimi K3—
Qwen3.6-27B93.5%
Source
Not directly comparable
SuperGPQA
Kimi K3—
Qwen3.6-27B66%
Source
Not directly comparable
C-Eval
Kimi K3—
Qwen3.6-27B91.4%
Source
Not directly comparable

Math

HMMT Feb 2025
Kimi K3—
Qwen3.6-27B93.8%
Source
Not directly comparable
HMMT Nov 2025
Kimi K3—
Qwen3.6-27B90.7%
Source
Not directly comparable
HMMT Feb 2026
Kimi K3—
Qwen3.6-27B84.3%
Source
Not directly comparable
MMAnswerBench
Kimi K3—
Qwen3.6-27B80.8%
Source
Not directly comparable
AIME26
Kimi K3—
Qwen3.6-27B94.1%
Source
Not directly comparable

Multimodal

OfficeQA Pro
Kimi K363.3%
Source
Qwen3.6-27B—
Not directly comparable
MMMU-Pro
Kimi K381.6%
Source
Qwen3.6-27B75.8%
Source
Kimi K3 leads this result
MMMU-Pro w/ Python
Kimi K383.4%
Source
Qwen3.6-27B—
Not directly comparable
CharXiv w/o tools
Kimi K384.8%
Source
Qwen3.6-27B—
Not directly comparable
CharXiv
Kimi K391.3%
Source
Qwen3.6-27B78.4%
Source
Kimi K3 leads this result
MathVision
Kimi K394.3%
Source
Qwen3.6-27B—
Not directly comparable
MathVision w/ Python
Kimi K397.8%
Source
Qwen3.6-27B—
Not directly comparable
BabyVision w/ Python
Kimi K385.7%
Source
Qwen3.6-27B—
Not directly comparable
ZeroBench
Kimi K323.0%
Source
Qwen3.6-27B—
Not directly comparable
ZeroBench w/ Python
Kimi K341.0%
Source
Qwen3.6-27B—
Not directly comparable
WorldVQA ForceAnswer
Kimi K351.0%
Source
Qwen3.6-27B—
Not directly comparable
OmniDocBench
Kimi K391.1%
Source
Qwen3.6-27B—
Not directly comparable
PerceptionBench
Kimi K358.5%
Source
Qwen3.6-27B—
Not directly comparable
MMMU
Kimi K3—
Qwen3.6-27B82.9%
Source
Not directly comparable
RealWorldQA
Kimi K3—
Qwen3.6-27B84.1%
Source
Not directly comparable
DynaMath
Kimi K3—
Qwen3.6-27B85.6%
Source
Not directly comparable
MStar
Kimi K3—
Qwen3.6-27B81.4%
Source
Not directly comparable
SimpleVQA
Kimi K3—
Qwen3.6-27B56.1%
Source
Not directly comparable
CC-OCR
Kimi K3—
Qwen3.6-27B81.2%
Source
Not directly comparable
CountBench
Kimi K3—
Qwen3.6-27B97.8%
Source
Not directly comparable
RefCOCO (avg)
Kimi K3—
Qwen3.6-27B92.5%
Source
Not directly comparable
ERQA
Kimi K3—
Qwen3.6-27B62.5%
Source
Not directly comparable
Video-MME (with subtitle)
Kimi K3—
Qwen3.6-27B87.7%
Source
Not directly comparable
VideoMMMU
Kimi K3—
Qwen3.6-27B84.4%
Source
Not directly comparable
MLVU (M-Avg)
Kimi K3—
Qwen3.6-27B86.6%
Source
Not directly comparable
V*
Kimi K3—
Qwen3.6-27B94.7%
Source
Not directly comparable

Frequently asked questions

Which is better, Kimi K3 or Qwen3.6-27B?

Kimi K3 has the higher public score, 80.53 versus 53.71, and the 90% score intervals do not overlap. That is the clearest overall quality signal in the current public data.

Which is better for coding, Kimi K3 or Qwen3.6-27B?

The published evidence does not provide a shared weighted coding basis for both models, so BenchLM does not name a coding winner.

Which is better for agentic tasks, Kimi K3 or Qwen3.6-27B?

The current agentic tasks averages use different weighted benchmark sets, so BenchLM does not name a winner from them. Read the shared benchmark rows directly and test the models on the same task set.

Which costs less, Kimi K3 or Qwen3.6-27B?

Both models do not have comparable published API token rates, so this page does not name a universal price winner.

Which has the larger context window, Kimi K3 or Qwen3.6-27B?

Kimi K3 has the larger documented context window: 1.05M, compared with 262K.

Related comparisons

Compare API pricing Read the methodology Open the model selector

Last updated August 19, 2026

Watch Kimi K3 vs Qwen3.6-27B

One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.

Read a sample issue

Join 2,000+ readers.

Kimi K3 vs Qwen3.6-27B

Which one for your work

Long documents

Coding work

Agentic work

Chat turn cost

Cache-heavy agent loop cost

Repository review cost

What is actually comparable

Category results, on a stated basis

Agentic

Knowledge

Multimodal

Coding

Reasoning

Math

Multilingual

Instruction following

Shape of the matched evidence

What each workload costs

Chat turn

Repository review

Cache-heavy agent loop

Specification differences

Context window

API model ID

Cached-input rate

Documented inputs

Documented outputs

Provider availability

Reasoning profile

Weight access

License

Release date

Self-host vs API cost

Benchmark evidence

Agentic

Coding

Knowledge

Math

Multimodal

Frequently asked questions

Which is better, Kimi K3 or Qwen3.6-27B?

Which is better for coding, Kimi K3 or Qwen3.6-27B?

Which is better for agentic tasks, Kimi K3 or Qwen3.6-27B?

Which costs less, Kimi K3 or Qwen3.6-27B?

Which has the larger context window, Kimi K3 or Qwen3.6-27B?

Related comparisons

Watch Kimi K3 vs Qwen3.6-27B