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 the free Radar Brief
BenchLM
OpenAI logo
Model A
GPT-5.4

OpenAI

73.56/100

Supported · Public rank #12

90% interval 70.4–76.7

GPT-5.4 vs MiniMax M3

Updated August 30, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.

MiniMax logo
Model B
MiniMax M3

MiniMax

68.73/100

Supported · Public rank #21

90% interval 63.5–74.0

Decision reading

GPT-5.4 has the higher public score estimate, 73.56 versus 68.73, but the 90% score intervals overlap. Treat that as a lead, not a settled winner.

9 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.

  • Agentic work

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

    GPT-5.4

    GPT-5.4 leads on the same 3 weighted benchmark rows.

    Confidence: stronger

  • Long documents

    Prompts that approach the documented context limit

    GPT-5.4

    GPT-5.4 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

Show secondary and unsupported calls

  • 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

  • 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

    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
9
GPT-5.4 only
28
MiniMax M3 only
13
Like-for-like categories
1 / 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

Like-for-like
GPT-5.4
77.2
MiniMax M3
72.3
Weighted basis
3 vs 3 rows
Reading
GPT-5.4 leads

Coding

Directional only
GPT-5.4
57.7
MiniMax M3
72.2
Weighted basis
1 vs 2 rows
Reading
Directional only

Multimodal

Directional only
GPT-5.4
73.2
MiniMax M3
64.9
Weighted basis
3 vs 2 rows
Reading
Directional only

Reasoning

Not comparable
GPT-5.4
74.0
MiniMax M3
Not measured
Weighted basis
1 vs 0 rows
Reading
Not comparable

Knowledge

Not comparable
GPT-5.4
57.6
MiniMax M3
Not measured
Weighted basis
2 vs 0 rows
Reading
Not comparable

Math

Not comparable
GPT-5.4
42.5
MiniMax M3
85.7
Weighted basis
2 vs 1 rows
Reading
Not comparable

Multilingual

Not comparable
GPT-5.4
Not measured
MiniMax M3
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Instruction following

Not comparable
GPT-5.4
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

GPT-5.4
$0.01
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

GPT-5.4
$0.17
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

GPT-5.4
$0.25
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.

GPT-5.4

1.05M

OpenAI pricing

MiniMax M3

1M

Cached-input rate

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

GPT-5.4

$0.25 per 1M cached input tokens

OpenAI pricing

MiniMax M3

$0.06 per 1M cached input tokens

Documented inputs

GPT-5.4

Not sourced

MiniMax M3

Not sourced

Documented outputs

GPT-5.4

Not sourced

MiniMax M3

Not sourced

Provider availability

GPT-5.4

Not sourced

MiniMax M3

Not sourced

Reasoning profile

GPT-5.4

Reasoning

MiniMax M3

Non-Reasoning

Weight access

GPT-5.4

Proprietary

MiniMax M3

Open Weight

License

GPT-5.4

Proprietary

MiniMax M3

Open Weight

Release date

GPT-5.4

2026-03-05

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
GPT-5.4 has the higher public score estimate, 73.56 versus 68.73, but the 90% score intervals overlap.
Workload cost
Repository review: $0.17 vs $0.0186. Cache-heavy agent loop: $0.25 vs $0.03.
Context tradeoff
GPT-5.4 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 evidence50 rows

Agentic

  • Terminal-Bench 2.0

    GPT-5.475.1%
    Source
    MiniMax M366%
    Source

    GPT-5.4 leads this result

  • CyberGym

    GPT-5.479.0%
    Source
    MiniMax M3—

    Not directly comparable

  • BrowseComp

    GPT-5.482.7%
    Source
    MiniMax M383.5%
    Source

    MiniMax M3 leads this result

  • OSWorld-Verified

    GPT-5.475%
    Source
    MiniMax M370.1%
    Source

    GPT-5.4 leads this result

  • MCP Atlas

    GPT-5.470.6%
    Source
    MiniMax M374.2%
    Source

    MiniMax M3 leads this result

  • Toolathlon

    GPT-5.454.6%
    Source
    MiniMax M3—

    Not directly comparable

  • τ²-bench results

    GPT-5.498.9%
    Source
    MiniMax M3—

    Not directly comparable

  • Claw-Eval

    GPT-5.460.3%
    Source
    MiniMax M374.5%
    Source

    MiniMax M3 leads this result

  • DeepSearchQA

    GPT-5.473.6%
    Source
    MiniMax M3—

    Not directly comparable

  • Gert Labs

    GPT-5.464.89%
    Source
    MiniMax M3—

    Not directly comparable

  • ResearchClawBench

    Shared source
    GPT-5.415.3%
    MiniMax M319.8%

    MiniMax M3 leads this result

  • JobBench

    GPT-5.438.9%
    Source
    MiniMax M3—

    Not directly comparable

  • ExploitGym

    GPT-5.46.0%
    Source
    MiniMax M3—

    Not directly comparable

  • BankerToolBench

    GPT-5.4—
    MiniMax M376.1%
    Source

    Not directly comparable

  • OSWorld 2.0

    GPT-5.4—
    MiniMax M34.6%
    Source

    Not directly comparable

Coding

  • LiveCodeBench Pro

    GPT-5.487.5%
    Source
    MiniMax M3—

    Not directly comparable

  • SWE-bench Pro

    GPT-5.457.7%
    Source
    MiniMax M359%
    Source

    MiniMax M3 leads this result

  • React Native Evals

    GPT-5.485.3%
    Source
    MiniMax M3—

    Not directly comparable

  • Vibe Code Bench

    GPT-5.467.42%
    Source
    MiniMax M3—

    Not directly comparable

  • SWE-bench Verified

    GPT-5.4—
    MiniMax M380.5%
    Source

    Not directly comparable

  • Terminal-Bench 2.0

    GPT-5.4—
    MiniMax M366.0%
    Source

    Not directly comparable

  • NL2Repo

    GPT-5.4—
    MiniMax M342.1%
    Source

    Not directly comparable

  • VIBE V2

    GPT-5.4—
    MiniMax M350.1%
    Source

    Not directly comparable

  • SVG-Bench

    GPT-5.4—
    MiniMax M363.7%
    Source

    Not directly comparable

  • KernelBench Hard

    GPT-5.4—
    MiniMax M328.8%
    Source

    Not directly comparable

  • OpenHarmony Bench

    GPT-5.4—
    MiniMax M348.4%
    Source

    Not directly comparable

Reasoning

  • ARC-AGI-2

    GPT-5.474.0%
    Source
    MiniMax M3—

    Not directly comparable

  • ARC-AGI-3

    GPT-5.40.2%
    Source
    MiniMax M3—

    Not directly comparable

Knowledge

  • GPQA

    GPT-5.492.8%
    Source
    MiniMax M3—

    Not directly comparable

  • HLE

    GPT-5.452.1%
    Source
    MiniMax M3—

    Not directly comparable

  • HLE w/o tools

    GPT-5.439.8%
    Source
    MiniMax M3—

    Not directly comparable

  • GPQA-D

    GPT-5.492.8%
    Source
    MiniMax M3—

    Not directly comparable

  • HealthBench Hard

    GPT-5.440.1%
    Source
    MiniMax M3—

    Not directly comparable

  • MedXpertQA (Text)

    GPT-5.459.6%
    Source
    MiniMax M3—

    Not directly comparable

  • HealthBench Professional

    GPT-5.448.1%
    Source
    MiniMax M3—

    Not directly comparable

Math

  • FrontierMath v2 (Tiers 1-3)

    GPT-5.447.600%
    Source
    MiniMax M3—

    Not directly comparable

  • FrontierMath v2 (Tier 4)

    GPT-5.427.100%
    Source
    MiniMax M3—

    Not directly comparable

  • USAMO 2026

    GPT-5.4—
    MiniMax M385.7%
    Source

    Not directly comparable

Multimodal

  • MMMU-Pro

    GPT-5.481.2%
    Source
    MiniMax M378.1%
    Source

    GPT-5.4 leads this result

  • OfficeQA Pro

    GPT-5.453.2%
    Source
    MiniMax M345.1%
    Source

    GPT-5.4 leads this result

  • MMMU-Pro w/ Python

    GPT-5.482.1%
    Source
    MiniMax M3—

    Not directly comparable

  • CharXiv

    GPT-5.482.8%
    Source
    MiniMax M3—

    Not directly comparable

  • ERQA

    GPT-5.465.4%
    Source
    MiniMax M3—

    Not directly comparable

  • SimpleVQA

    GPT-5.461.1%
    Source
    MiniMax M3—

    Not directly comparable

  • ScreenSpot Pro

    GPT-5.485.4%
    Source
    MiniMax M3—

    Not directly comparable

  • ZeroBench

    GPT-5.441.0%
    Source
    MiniMax M3—

    Not directly comparable

  • MedXpertQA (MM)

    GPT-5.477.1%
    Source
    MiniMax M3—

    Not directly comparable

  • OmniDocBench 1.5

    GPT-5.4—
    MiniMax M391.6%
    Source

    Not directly comparable

  • VideoMMMU

    GPT-5.4—
    MiniMax M384.6%
    Source

    Not directly comparable

  • Video-MME (with subtitle)

    GPT-5.4—
    MiniMax M385.4%
    Source

    Not directly comparable

Frequently asked questions

Which is better, GPT-5.4 or MiniMax M3?

GPT-5.4 has the higher public score estimate, 73.56 versus 68.73, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.

Which is better for coding, GPT-5.4 or MiniMax M3?

The current coding 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 is better for agentic tasks, GPT-5.4 or MiniMax M3?

GPT-5.4 leads the like-for-like agentic tasks comparison across 3 shared weighted benchmark rows.

Which costs less, GPT-5.4 or MiniMax M3?

For the stated presets, chat costs $0.01 on GPT-5.4 and $0.0009 on MiniMax M3; repository review costs $0.17 and $0.0186; the cache-heavy agent loop costs $0.25 and $0.03. Costs use the listed standard API rates.

Which has the larger context window, GPT-5.4 or MiniMax M3?

GPT-5.4 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 30, 2026

Watch GPT-5.4 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.