Skip to main content
Radar

Keep up with the models you depend on. Follow price changes, retirements, and API updates.Follow the models you depend on.

Follow model changes
BenchLM
OpenAI logo
Model A
GPT Audio 1.5

OpenAI

—

Evidence status unavailable

90% interval unavailable

GPT Audio 1.5 vs Kimi K3

Updated September 10, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.

Moonshot AI logo
Model B
Kimi K3

Moonshot AI

74.82/100

Supported · Public rank #8

90% interval 71.4–78.2

Decision reading

The public evidence has no benchmark result shared by both models, so it does not support a quality verdict. Use the documented cost, context, and runtime rows instead.

0 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.

Share or export

Share on XLinkedInSocial cardCSVJSON

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

    GPT Audio 1.5

    GPT Audio 1.5 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

    GPT Audio 1.5

    GPT Audio 1.5 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

    GPT Audio 1.5 is not ranked on the public lane for coding, so no winner is named for coding.

    Confidence: limited

  • Agentic work

    Tool use, computer use, and multi-step task completion

    Not enough matched evidence

    GPT Audio 1.5 is not ranked on the public lane for agentic, so no winner is named for agentic.

    Confidence: limited

  • Cache-heavy agent loop cost

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

    Not enough matched evidence

    The page does not recommend a cost winner because at least one model cannot fit the stated workload in one request. GPT Audio 1.5 does not fit this workload in one request. GPT Audio 1.5 has no published cached-input rate, so cached tokens use its listed input rate.

    Confidence: rate-fallback

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
0
GPT Audio 1.5 only
0
Kimi K3 only
43
Like-for-like categories
0 / 8

Category results, on a stated basis

Each row shows the public-lane category score for both models: the BenchAlign lane for agentic, coding, and knowledge, and the weighted public rows elsewhere, with the same rank each model page reports. A row is like-for-like only when both scores rest on Supported evidence or the same weighted set. Directional and not-comparable rows remain visible, but they never receive a winner.

Agentic

Not comparable
GPT Audio 1.5
Not ranked
Kimi K3
71.9
Supported · #4/152
Basis
BenchAlign lane · 0 vs 11 public rows
Reading
Not comparable

Coding

Not comparable
GPT Audio 1.5
Not ranked
Kimi K3
68.0
Supported · #6/151
Basis
BenchAlign lane · 0 vs 13 public rows
Reading
Not comparable

Reasoning

Not comparable
GPT Audio 1.5
Not ranked
Kimi K3
78.5
#3/20
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Not comparable

Knowledge

Not comparable
GPT Audio 1.5
Not ranked
Kimi K3
72.0
Supported · #8/183
Basis
BenchAlign lane · 0 vs 6 public rows
Reading
Not comparable

Math

Not comparable
GPT Audio 1.5
Not ranked
Kimi K3
Not ranked
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Not comparable

Multilingual

Not comparable
GPT Audio 1.5
Not ranked
Kimi K3
Not ranked
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Not comparable

Multimodal

Not comparable
GPT Audio 1.5
Not ranked
Kimi K3
89.5
#1/48
Basis
Provisional lane · 0 vs 3 weighted rows
Reading
Not comparable

Instruction following

Not comparable
GPT Audio 1.5
Not ranked
Kimi K3
Not ranked
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Not comparable

Ranks count the models scored in each category’s lane, so the agentic, coding, and knowledge denominators (BenchAlign) differ from the provisional-lane categories. Unranked scores sit on the lane’s scale but never name a winner. Like-for-like rows within 0.5 points read as a practical tie.

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.

A shared-evidence shape is not available.

BenchLM does not draw a radar or infer missing axes when the matched evidence is too sparse.

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

GPT Audio 1.5
$0.0075
Fits in one request
Kimi K3
$0.0105
Fits in one request

GPT Audio 1.5 has the lower modeled cost

Costs use the listed standard API rates.

Repository review

50K fresh input + 3K output tokens

GPT Audio 1.5
$0.155
Fits in one request
Kimi K3
$0.195
Fits in one request

GPT Audio 1.5 has the lower modeled cost

Costs use the listed standard API rates.

Cache-heavy agent loop

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

GPT Audio 1.5
$0.65
Does not fit in one request
Cached input priced at the published list-input rate
Kimi K3
$0.27
Fits in one request

GPT Audio 1.5 does not fit this workload in one request. GPT Audio 1.5 has no published cached-input rate, so cached tokens use its listed input rate.

Specification differences

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

Cached-input rate

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

GPT Audio 1.5

Not published

OpenAI model documentation

Kimi K3

$0.3 per 1M cached input tokens

Reasoning profile

GPT Audio 1.5

Non-Reasoning

Kimi K3

Reasoning

Weight access

GPT Audio 1.5

Proprietary

Kimi K3

Pending

License

GPT Audio 1.5

Proprietary

Kimi K3

Pending

Release date

GPT Audio 1.5

2026-04-23

Kimi K3

2026-07-16

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
The public evidence has no benchmark result shared by both models, so it does not support a quality verdict.
Workload cost
Repository review: $0.155 vs $0.195. Cache-heavy agent loop: $0.65 vs $0.27.
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

    GPT Audio 1.5—
    Kimi K388.3%
    Source

    Not directly comparable

  • BrowseComp

    GPT Audio 1.5—
    Kimi K391.2%
    Source

    Not directly comparable

  • DeepSearchQA

    GPT Audio 1.5—
    Kimi K395.0%
    Source

    Not directly comparable

  • Toolathlon-Verified

    GPT Audio 1.5—
    Kimi K373.2%
    Source

    Not directly comparable

  • MCP Atlas

    GPT Audio 1.5—
    Kimi K384.2%
    Source

    Not directly comparable

  • AutomationBench

    GPT Audio 1.5—
    Kimi K330.8%
    Source

    Not directly comparable

  • JobBench

    GPT Audio 1.5—
    Kimi K352.9%
    Source

    Not directly comparable

  • APEX-Agents

    GPT Audio 1.5—
    Kimi K337.6%
    Source

    Not directly comparable

  • SpreadsheetBench 2

    GPT Audio 1.5—
    Kimi K334.8%
    Source

    Not directly comparable

  • DECK-Bench

    GPT Audio 1.5—
    Kimi K373.5%
    Source

    Not directly comparable

  • Terminal-Bench 2.1 (Vals)

    GPT Audio 1.5—
    Kimi K380.9%
    Source

    Not directly comparable

Coding

  • DeepSWE

    GPT Audio 1.5—
    Kimi K367.5%
    Source

    Not directly comparable

  • cursorBench32

    GPT Audio 1.5—
    Kimi K360.8%
    Source

    Not directly comparable

  • FrontierSWE

    GPT Audio 1.5—
    Kimi K381.2%
    Source

    Not directly comparable

  • ProgramBench

    GPT Audio 1.5—
    Kimi K377.8%
    Source

    Not directly comparable

  • Kimi Code Bench v2

    GPT Audio 1.5—
    Kimi K372.9%
    Source

    Not directly comparable

  • sweMarathon

    GPT Audio 1.5—
    Kimi K342%
    Source

    Not directly comparable

  • PostTrain Bench

    GPT Audio 1.5—
    Kimi K336.6%
    Source

    Not directly comparable

  • MLS-Bench Lite

    GPT Audio 1.5—
    Kimi K348.3%
    Source

    Not directly comparable

  • VulcanBench v3

    GPT Audio 1.5—
    Kimi K373.7%
    Source

    Not directly comparable

  • OpenHarmony Bench

    GPT Audio 1.5—
    Kimi K357.3%
    Source

    Not directly comparable

  • FrontierSWE v2

    GPT Audio 1.5—
    Kimi K325.9%
    Source

    Not directly comparable

  • LiveCodeBench (Vals)

    GPT Audio 1.5—
    Kimi K387.2%
    Source

    Not directly comparable

  • SWE-bench (Vals)

    GPT Audio 1.5—
    Kimi K393.4%
    Source

    Not directly comparable

Knowledge

  • GPQA

    GPT Audio 1.5—
    Kimi K393.5%
    Source

    Not directly comparable

  • GPQA-D

    GPT Audio 1.5—
    Kimi K393.5%
    Source

    Not directly comparable

  • HLE

    GPT Audio 1.5—
    Kimi K356%
    Source

    Not directly comparable

  • HLE w/o tools

    GPT Audio 1.5—
    Kimi K343.5%
    Source

    Not directly comparable

  • GPQA Diamond (Vals)

    GPT Audio 1.5—
    Kimi K392.9%
    Source

    Not directly comparable

  • MMLU-Pro (Vals)

    GPT Audio 1.5—
    Kimi K388.0%
    Source

    Not directly comparable

Multimodal

  • OfficeQA Pro

    GPT Audio 1.5—
    Kimi K363.3%
    Source

    Not directly comparable

  • MMMU-Pro

    GPT Audio 1.5—
    Kimi K381.6%
    Source

    Not directly comparable

  • MMMU-Pro w/ Python

    GPT Audio 1.5—
    Kimi K383.4%
    Source

    Not directly comparable

  • CharXiv w/o tools

    GPT Audio 1.5—
    Kimi K384.8%
    Source

    Not directly comparable

  • CharXiv

    GPT Audio 1.5—
    Kimi K391.3%
    Source

    Not directly comparable

  • MathVision

    GPT Audio 1.5—
    Kimi K394.3%
    Source

    Not directly comparable

  • MathVision w/ Python

    GPT Audio 1.5—
    Kimi K397.8%
    Source

    Not directly comparable

  • BabyVision w/ Python

    GPT Audio 1.5—
    Kimi K385.7%
    Source

    Not directly comparable

  • ZeroBench

    GPT Audio 1.5—
    Kimi K323.0%
    Source

    Not directly comparable

  • ZeroBench w/ Python

    GPT Audio 1.5—
    Kimi K341.0%
    Source

    Not directly comparable

  • WorldVQA ForceAnswer

    GPT Audio 1.5—
    Kimi K351.0%
    Source

    Not directly comparable

  • OmniDocBench

    GPT Audio 1.5—
    Kimi K391.1%
    Source

    Not directly comparable

  • PerceptionBench

    GPT Audio 1.5—
    Kimi K358.5%
    Source

    Not directly comparable

Frequently asked questions

Which is better, GPT Audio 1.5 or Kimi K3?

The public evidence has no benchmark result shared by both models, so it does not support a quality verdict. The page therefore keeps the decision tied to the specific documented workload.

Which is better for coding, GPT Audio 1.5 or Kimi K3?

GPT Audio 1.5 is not ranked on the public lane for coding, so no winner is named for coding.

Which is better for agentic tasks, GPT Audio 1.5 or Kimi K3?

GPT Audio 1.5 is not ranked on the public lane for agentic tasks, so no winner is named for agentic tasks.

Which costs less, GPT Audio 1.5 or Kimi K3?

For the stated presets, chat costs $0.0075 on GPT Audio 1.5 and $0.0105 on Kimi K3; repository review costs $0.155 and $0.195; the cache-heavy agent loop costs $0.65 and $0.27. GPT Audio 1.5 does not fit this workload in one request. GPT Audio 1.5 has no published cached-input rate, so cached tokens use its listed input rate.

Which has the larger context window, GPT Audio 1.5 or Kimi K3?

Kimi K3 has the larger documented context window: 1.05M, compared with 128K.

Related comparisons

Compare API pricing Read the methodology Open the model selector

Last updated September 10, 2026

Watch GPT Audio 1.5 vs Kimi K3

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

Read a sample issue

Join 2,000+ readers.