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 MiniMax M3

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

Model B
MiniMax M3

MiniMax

68.8/100

Supported · Public rank #19

90% interval 63.8–73.8

Decision reading

Kimi K3 has the higher public score, 80.53 versus 68.76, and the 90% score intervals do not overlap.

6 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

  • Chat turn cost

    1K fresh input + 500 output tokens

    MiniMax M3

    MiniMax M3 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.

    Confidence: listed-rates

  • Cache-heavy agent loop cost

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

    MiniMax M3

    MiniMax M3 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.

    Confidence: listed-rates

Show secondary and unsupported calls

  • Repository review cost

    50K fresh input + 3K output tokens

    MiniMax M3

    MiniMax M3 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.

    Confidence: listed-rates

  • 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

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
6
Kimi K3 only
34
MiniMax M3 only
16
Like-for-like categories
0 / 8

2 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
MiniMax M3
72.3
Weighted basis
2 vs 3 rows
Reading
Directional only

Multimodal

Directional only
Kimi K3
78.5
MiniMax M3
64.9
Weighted basis
3 vs 2 rows
Reading
Directional only

Coding

Not comparable
Kimi K3
Not measured
MiniMax M3
72.2
Weighted basis
0 vs 2 rows
Reading
Not comparable

Reasoning

Not comparable
Kimi K3
Not measured
MiniMax M3
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Knowledge

Not comparable
Kimi K3
61.0
MiniMax M3
Not measured
Weighted basis
2 vs 0 rows
Reading
Not comparable

Math

Not comparable
Kimi K3
Not measured
MiniMax M3
85.7
Weighted basis
0 vs 1 rows
Reading
Not comparable

Multilingual

Not comparable
Kimi K3
Not measured
MiniMax M3
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Instruction following

Not comparable
Kimi K3
Not measured
MiniMax M3
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
MiniMax M3
$0.0009
Fits in one request

MiniMax M3 has the lower modeled cost

Costs use the listed standard API rates.

Repository review

50K fresh input + 3K output tokens

Kimi K3
$0.195
Fits in one request
MiniMax M3
$0.0186
Fits in one request

MiniMax M3 has the lower modeled cost

Costs use the listed standard API rates.

Cache-heavy agent loop

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

Kimi K3
$0.27
Fits in one request
MiniMax M3
$0.03
Fits in one request

MiniMax M3 has the lower modeled cost

Costs use the listed standard API rates.

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

MiniMax M3

1M

API model ID

Kimi K3

Not sourced

MiniMax M3

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

MiniMax M3

$0.06 per 1M cached input tokens

Documented inputs

Kimi K3

Not sourced

MiniMax M3

Not sourced

Documented outputs

Kimi K3

Not sourced

MiniMax M3

Not sourced

Provider availability

Kimi K3

Not sourced

MiniMax M3

Not sourced

Reasoning profile

Kimi K3

Reasoning

MiniMax M3

Non-Reasoning

Weight access

Kimi K3

Pending

MiniMax M3

Open Weight

License

Kimi K3

Pending

MiniMax M3

Open Weight

Release date

Kimi K3

2026-07-16

MiniMax M3

2026-06-01

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 68.76, and the 90% score intervals do not overlap.
Workload cost
Repository review: $0.195 vs $0.0186. Cache-heavy agent loop: $0.27 vs $0.03.
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 evidence56 rows

Agentic

  • Terminal-Bench 2.0

    Kimi K388.3%
    Source
    MiniMax M366%
    Source

    Kimi K3 leads this result

  • BrowseComp

    Kimi K391.2%
    Source
    MiniMax M383.5%
    Source

    Kimi K3 leads this result

  • DeepSearchQA

    Kimi K395.0%
    Source
    MiniMax M3—

    Not directly comparable

  • Toolathlon-Verified

    Kimi K373.2%
    Source
    MiniMax M3—

    Not directly comparable

  • MCP Atlas

    Kimi K384.2%
    Source
    MiniMax M374.2%
    Source

    Kimi K3 leads this result

  • AutomationBench

    Kimi K330.8%
    Source
    MiniMax M3—

    Not directly comparable

  • JobBench

    Kimi K352.9%
    Source
    MiniMax M3—

    Not directly comparable

  • APEX-Agents

    Kimi K337.6%
    Source
    MiniMax M3—

    Not directly comparable

  • SpreadsheetBench 2

    Kimi K334.8%
    Source
    MiniMax M3—

    Not directly comparable

  • DECK-Bench

    Kimi K373.5%
    Source
    MiniMax M3—

    Not directly comparable

  • OSWorld-Verified

    Kimi K3—
    MiniMax M370.1%
    Source

    Not directly comparable

  • Claw-Eval

    Kimi K3—
    MiniMax M374.5%
    Source

    Not directly comparable

  • BankerToolBench

    Kimi K3—
    MiniMax M376.1%
    Source

    Not directly comparable

  • ResearchClawBench

    Kimi K3—
    MiniMax M319.8%
    Source

    Not directly comparable

  • OSWorld 2.0

    Kimi K3—
    MiniMax M34.6%
    Source

    Not directly comparable

Coding

  • deepSwe

    Kimi K367.5%
    Source
    MiniMax M3—

    Not directly comparable

  • cursorBench32

    Kimi K360.8%
    Source
    MiniMax M3—

    Not directly comparable

  • FrontierSWE

    Kimi K381.2%
    Source
    MiniMax M3—

    Not directly comparable

  • ProgramBench

    Kimi K377.8%
    Source
    MiniMax M3—

    Not directly comparable

  • Kimi Code Bench v2

    Kimi K372.9%
    Source
    MiniMax M3—

    Not directly comparable

  • sweMarathon

    Kimi K342%
    Source
    MiniMax M3—

    Not directly comparable

  • PostTrain Bench

    Kimi K336.6%
    Source
    MiniMax M3—

    Not directly comparable

  • MLS-Bench Lite

    Kimi K348.3%
    Source
    MiniMax M3—

    Not directly comparable

  • VulcanBench v3

    Kimi K373.9%
    Source
    MiniMax M3—

    Not directly comparable

  • APEX-SWE

    Kimi K348.0%
    Source
    MiniMax M3—

    Not directly comparable

  • EEBench

    Shared source
    Kimi K338.3%
    MiniMax M39.2%

    Kimi K3 leads this result

  • InferenceEval

    Kimi K339.8%
    Source
    MiniMax M3—

    Not directly comparable

  • KernelBench Internal

    Kimi K368.5%
    Source
    MiniMax M3—

    Not directly comparable

  • SWE-bench Verified

    Kimi K3—
    MiniMax M380.5%
    Source

    Not directly comparable

  • SWE-bench Pro

    Kimi K3—
    MiniMax M359%
    Source

    Not directly comparable

  • Terminal-Bench 2.0

    Kimi K3—
    MiniMax M366.0%
    Source

    Not directly comparable

  • NL2Repo

    Kimi K3—
    MiniMax M342.1%
    Source

    Not directly comparable

  • VIBE V2

    Kimi K3—
    MiniMax M350.1%
    Source

    Not directly comparable

  • SVG-Bench

    Kimi K3—
    MiniMax M363.7%
    Source

    Not directly comparable

  • KernelBench Hard

    Kimi K3—
    MiniMax M328.8%
    Source

    Not directly comparable

Knowledge

  • GPQA

    Kimi K393.5%
    Source
    MiniMax M3—

    Not directly comparable

  • GPQA-D

    Kimi K393.5%
    Source
    MiniMax M3—

    Not directly comparable

  • HLE

    Kimi K356%
    Source
    MiniMax M3—

    Not directly comparable

  • HLE w/o tools

    Kimi K343.5%
    Source
    MiniMax M3—

    Not directly comparable

Math

  • USAMO 2026

    Kimi K3—
    MiniMax M385.7%
    Source

    Not directly comparable

Multimodal

  • OfficeQA Pro

    Kimi K363.3%
    Source
    MiniMax M345.1%
    Source

    Kimi K3 leads this result

  • MMMU-Pro

    Kimi K381.6%
    Source
    MiniMax M378.1%
    Source

    Kimi K3 leads this result

  • MMMU-Pro w/ Python

    Kimi K383.4%
    Source
    MiniMax M3—

    Not directly comparable

  • CharXiv w/o tools

    Kimi K384.8%
    Source
    MiniMax M3—

    Not directly comparable

  • CharXiv

    Kimi K391.3%
    Source
    MiniMax M3—

    Not directly comparable

  • MathVision

    Kimi K394.3%
    Source
    MiniMax M3—

    Not directly comparable

  • MathVision w/ Python

    Kimi K397.8%
    Source
    MiniMax M3—

    Not directly comparable

  • BabyVision w/ Python

    Kimi K385.7%
    Source
    MiniMax M3—

    Not directly comparable

  • ZeroBench

    Kimi K323.0%
    Source
    MiniMax M3—

    Not directly comparable

  • ZeroBench w/ Python

    Kimi K341.0%
    Source
    MiniMax M3—

    Not directly comparable

  • WorldVQA ForceAnswer

    Kimi K351.0%
    Source
    MiniMax M3—

    Not directly comparable

  • OmniDocBench

    Kimi K391.1%
    Source
    MiniMax M3—

    Not directly comparable

  • PerceptionBench

    Kimi K358.5%
    Source
    MiniMax M3—

    Not directly comparable

  • OmniDocBench 1.5

    Kimi K3—
    MiniMax M391.6%
    Source

    Not directly comparable

  • VideoMMMU

    Kimi K3—
    MiniMax M384.6%
    Source

    Not directly comparable

  • Video-MME (with subtitle)

    Kimi K3—
    MiniMax M385.4%
    Source

    Not directly comparable

Frequently asked questions

Which is better, Kimi K3 or MiniMax M3?

Kimi K3 has the higher public score, 80.53 versus 68.76, 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 MiniMax M3?

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 MiniMax M3?

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 MiniMax M3?

For the stated presets, chat costs $0.0105 on Kimi K3 and $0.0009 on MiniMax M3; repository review costs $0.195 and $0.0186; the cache-heavy agent loop costs $0.27 and $0.03. Costs use the listed standard API rates.

Which has the larger context window, Kimi K3 or MiniMax M3?

Kimi K3 has the larger documented context window: 1.05M, compared with 1M.

Related comparisons

Compare API pricing Read the methodology Open the model selector

Last updated August 19, 2026

Watch Kimi K3 vs MiniMax M3

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

Read a sample issue

Join 2,000+ readers.