Skip to main content
Radar

Five or fewer confirmed AI changes, with original sources, on mornings when something changed.A free source-linked morning brief.

Start free brief
BenchLM
Model A
Kimi K3

Moonshot AI

80.5/100

Supported · Public rank #5

90% interval 77.7–83.4

Kimi K3 vs SWE-1.7

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

Model B
SWE-1.7

Cognition

—

Evidence status unavailable

90% interval unavailable

Decision reading

At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner.

1 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.

Shared results
1
Kimi K3 only
39
SWE-1.7 only
3
Like-for-like categories
0 / 8

1 category uses 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
SWE-1.7
81.5
Weighted basis
2 vs 1 rows
Reading
Directional only

Coding

Not comparable
Kimi K3
Not measured
SWE-1.7
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Reasoning

Not comparable
Kimi K3
Not measured
SWE-1.7
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Knowledge

Not comparable
Kimi K3
61.0
SWE-1.7
Not measured
Weighted basis
2 vs 0 rows
Reading
Not comparable

Math

Not comparable
Kimi K3
Not measured
SWE-1.7
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Multilingual

Not comparable
Kimi K3
Not measured
SWE-1.7
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Multimodal

Not comparable
Kimi K3
78.5
SWE-1.7
Not measured
Weighted basis
3 vs 0 rows
Reading
Not comparable

Instruction following

Not comparable
Kimi K3
Not measured
SWE-1.7
Not measured
Weighted basis
0 vs 0 rows
Reading
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.

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
SWE-1.7
API rate not published
Fits in one request

SWE-1.7 has no comparable published API token rate.

Repository review

50K fresh input + 3K output tokens

Kimi K3
$0.195
Fits in one request
SWE-1.7
API rate not published
Fits in one request

SWE-1.7 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
SWE-1.7
API rate not published
Fits in one request
Cached-input rate unavailable

SWE-1.7 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.

Context window

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

Kimi K3

1.05M

SWE-1.7

256K

API model ID

Kimi K3

Not sourced

SWE-1.7

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

SWE-1.7

No comparable hosted API rate

Documented inputs

Kimi K3

Not sourced

SWE-1.7

Not sourced

Documented outputs

Kimi K3

Not sourced

SWE-1.7

Not sourced

Provider availability

Kimi K3

Not sourced

SWE-1.7

Not sourced

Reasoning profile

Kimi K3

Reasoning

SWE-1.7

Reasoning

Weight access

Kimi K3

Pending

SWE-1.7

Proprietary

License

Kimi K3

Pending

SWE-1.7

Proprietary

Release date

Kimi K3

2026-07-16

SWE-1.7

2026-07-08

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
At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner.
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.

Benchmark evidence

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

Browse raw public benchmark evidence43 rows

Agentic

  • Terminal-Bench 2.0

    Kimi K388.3%
    Source
    SWE-1.781.5%
    Source

    Kimi K3 leads this result

  • BrowseComp

    Kimi K391.2%
    Source
    SWE-1.7—

    Not directly comparable

  • DeepSearchQA

    Kimi K395.0%
    Source
    SWE-1.7—

    Not directly comparable

  • Toolathlon-Verified

    Kimi K373.2%
    Source
    SWE-1.7—

    Not directly comparable

  • MCP Atlas

    Kimi K384.2%
    Source
    SWE-1.7—

    Not directly comparable

  • AutomationBench

    Kimi K330.8%
    Source
    SWE-1.7—

    Not directly comparable

  • JobBench

    Kimi K352.9%
    Source
    SWE-1.7—

    Not directly comparable

  • APEX-Agents

    Kimi K337.6%
    Source
    SWE-1.7—

    Not directly comparable

  • SpreadsheetBench 2

    Kimi K334.8%
    Source
    SWE-1.7—

    Not directly comparable

  • DECK-Bench

    Kimi K373.5%
    Source
    SWE-1.7—

    Not directly comparable

Coding

  • deepSwe

    Kimi K367.5%
    Source
    SWE-1.7—

    Not directly comparable

  • cursorBench32

    Kimi K360.8%
    Source
    SWE-1.7—

    Not directly comparable

  • FrontierSWE

    Kimi K381.2%
    Source
    SWE-1.7—

    Not directly comparable

  • ProgramBench

    Kimi K377.8%
    Source
    SWE-1.7—

    Not directly comparable

  • Kimi Code Bench v2

    Kimi K372.9%
    Source
    SWE-1.7—

    Not directly comparable

  • sweMarathon

    Kimi K342%
    Source
    SWE-1.7—

    Not directly comparable

  • PostTrain Bench

    Kimi K336.6%
    Source
    SWE-1.7—

    Not directly comparable

  • MLS-Bench Lite

    Kimi K348.3%
    Source
    SWE-1.7—

    Not directly comparable

  • VulcanBench v3

    Kimi K373.9%
    Source
    SWE-1.7—

    Not directly comparable

  • APEX-SWE

    Kimi K348.0%
    Source
    SWE-1.7—

    Not directly comparable

  • EEBench

    Kimi K338.3%
    Source
    SWE-1.7—

    Not directly comparable

  • InferenceEval

    Kimi K339.8%
    Source
    SWE-1.7—

    Not directly comparable

  • KernelBench Internal

    Kimi K368.5%
    Source
    SWE-1.7—

    Not directly comparable

  • FrontierCode 1.1 Main

    Kimi K3—
    SWE-1.742.3%
    Source

    Not directly comparable

  • Terminal-Bench 2.0

    Kimi K3—
    SWE-1.781.5%
    Source

    Not directly comparable

  • SWE Multilingual

    Kimi K3—
    SWE-1.777.8%
    Source

    Not directly comparable

Knowledge

  • GPQA

    Kimi K393.5%
    Source
    SWE-1.7—

    Not directly comparable

  • GPQA-D

    Kimi K393.5%
    Source
    SWE-1.7—

    Not directly comparable

  • HLE

    Kimi K356%
    Source
    SWE-1.7—

    Not directly comparable

  • HLE w/o tools

    Kimi K343.5%
    Source
    SWE-1.7—

    Not directly comparable

Multimodal

  • OfficeQA Pro

    Kimi K363.3%
    Source
    SWE-1.7—

    Not directly comparable

  • MMMU-Pro

    Kimi K381.6%
    Source
    SWE-1.7—

    Not directly comparable

  • MMMU-Pro w/ Python

    Kimi K383.4%
    Source
    SWE-1.7—

    Not directly comparable

  • CharXiv w/o tools

    Kimi K384.8%
    Source
    SWE-1.7—

    Not directly comparable

  • CharXiv

    Kimi K391.3%
    Source
    SWE-1.7—

    Not directly comparable

  • MathVision

    Kimi K394.3%
    Source
    SWE-1.7—

    Not directly comparable

  • MathVision w/ Python

    Kimi K397.8%
    Source
    SWE-1.7—

    Not directly comparable

  • BabyVision w/ Python

    Kimi K385.7%
    Source
    SWE-1.7—

    Not directly comparable

  • ZeroBench

    Kimi K323.0%
    Source
    SWE-1.7—

    Not directly comparable

  • ZeroBench w/ Python

    Kimi K341.0%
    Source
    SWE-1.7—

    Not directly comparable

  • WorldVQA ForceAnswer

    Kimi K351.0%
    Source
    SWE-1.7—

    Not directly comparable

  • OmniDocBench

    Kimi K391.1%
    Source
    SWE-1.7—

    Not directly comparable

  • PerceptionBench

    Kimi K358.5%
    Source
    SWE-1.7—

    Not directly comparable

Frequently asked questions

Which is better, Kimi K3 or SWE-1.7?

At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner. The page therefore keeps the decision tied to the specific documented workload.

Which is better for coding, Kimi K3 or SWE-1.7?

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 SWE-1.7?

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 SWE-1.7?

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 SWE-1.7?

Kimi K3 has the larger documented context window: 1.05M, compared with 256K.

Related comparisons

Compare API pricing Read the methodology Open the model selector

Last updated August 19, 2026

Watch Kimi K3 vs SWE-1.7

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

Read a sample issue

Join 2,000+ readers.