GPT-5.2 vs MiniMax M3

Updated October 14, 2026. Rank says GPT-5.2 is ahead. Price, access, and your workload can each overturn that. Public scores include evidence status and uncertainty.

Decision reading

GPT-5.2 has the higher public score estimate, 59.85 versus 55, but the 90% score intervals overlap. Treat that as a lead, not a settled winner. 8 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.

Model A

GPT-5.2

OpenAI

59.85/100

Supported · Public rank #56

90% interval 52.7–67.0

Model B

MiniMax M3

MiniMax

55/100

Supported · Public rank #68

90% interval 45.6–64.4

Shared results: 8
GPT-5.2 only: 23
MiniMax M3 only: 19
Like-for-like categories: 3 / 8

Supported: GPT-5.2 and MiniMax M3How the comparison works

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-5.2
GPT-5.2 has the higher public agentic point estimate, 44.8 to 39.6, with Supported evidence for both models. The conditional ranges do not establish rank confidence.
Confidence: limited
Long documents
Prompts that approach the documented context limit
MiniMax M3
MiniMax M3 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. GPT-5.2 has no published cached-input rate, so cached tokens use its listed input rate.
Confidence: rate-fallback
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
No clear pick
The like-for-like coding result is a practical tie on the public lane (within 0.5 points).
Confidence: limited

Which one for a specific job

Choose a job from the LLM Selector's task catalog to see the category row it rests on, under the same basis rules as the table below. A directional row stays directional; choosing a job never creates a winner.

Pick the job you would use them for

The same task catalog as the LLM Selector. Each job names the evidence surface it rests on; nothing here adds a new score.

36.9GPT-5.236.5MiniMax M3

Like-for-like · BenchAlign v5.8

GPT-5.2 has the higher coding point estimate. Conditional score ranges do not establish rank confidence.

Coding scores combine specific tasks and setups. Match the editor, harness, and effort to your workflow.

Open the coding leaderboard Shortlist models for this job

Same basis rules as the category table below

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.

OSWorld-VerifiedAgentic
Normalized gap 22.8
GPT-5.2:47.3%
MiniMax M3:70.1%
GPT-5.2 source MiniMax M3 source
BrowseCompAgentic
Normalized gap 17.7
GPT-5.2:65.8%
MiniMax M3:83.5%
GPT-5.2 source MiniMax M3 source
SWE-bench ProCoding
Normalized gap 3.4
GPT-5.2:55.6%
MiniMax M3:59%
GPT-5.2 source MiniMax M3 source
MMMU-ProMultimodal
Normalized gap 1.4
GPT-5.2:79.5%
MiniMax M3:78.1%
GPT-5.2 source MiniMax M3 source
SWE-bench VerifiedCoding
Normalized gap 0.5
GPT-5.2:80%
MiniMax M3:80.5%
GPT-5.2 source MiniMax M3 source

Bars run 0–100 on each benchmark’s normalized display scale

Category results, on a stated basis

Each row shows the public-lane category score for both models: the BenchAlign v5.8 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

Like-for-like

GPT-5.2: 44.8; Supported · #57/125
MiniMax M3: 39.6; Supported · #66/125
Basis: BenchAlign v5.8 lane · 4 vs 9 public rows
Reading: GPT-5.2 leads · intervals overlap

Coding

Like-for-like

GPT-5.2: 36.9; Supported · #72/149
MiniMax M3: 36.5; Supported · #73/149
Basis: BenchAlign v5.8 lane · 3 vs 10 public rows
Reading: Practical tie

Knowledge

Like-for-like

GPT-5.2: 59.6; Supported · #49/181
MiniMax M3: 49.6; Supported · #80/181
Basis: BenchAlign v5.8 lane · 1 vs 2 public rows
Reading: GPT-5.2 leads · intervals overlap

Multimodal

Directional only

GPT-5.2: 75.9; #22/54
MiniMax M3: 61.5; #43/54
Basis: Provisional lane · 2 vs 2 weighted rows
Reading: Directional only

Instruction following

Directional only

GPT-5.2: 92.6; #19/144
MiniMax M3: 93.8; #5/144
Basis: Provisional lane · 0 vs 0 weighted rows
Reading: Directional only

Reasoning

Not comparable

GPT-5.2: 64.3; Unranked · 4 rankable rows
MiniMax M3: 82.9; #5/29
Basis: Provisional lane · 1 vs 0 weighted rows
Reading: Not comparable

Multilingual

Not comparable

GPT-5.2: Not ranked
MiniMax M3: Not ranked
Basis: Provisional lane · 0 vs 0 weighted rows
Reading: Not comparable

Math

Not comparable

GPT-5.2: 71.1; Unranked · 3 rankable rows
MiniMax M3: Not ranked
Basis: Provisional lane · 2 vs 1 weighted rows
Reading: Not comparable

Category scores with the server-provided evidence basis for GPT-5.2 and MiniMax M3
Category	GPT-5.2	MiniMax M3	Basis	Reading
Agentic	44.8Supported · #57/125	39.6Supported · #66/125	Like-for-likeBenchAlign v5.8 lane · 4 vs 9 public rows	GPT-5.2 leads · intervals overlap
Coding	36.9Supported · #72/149	36.5Supported · #73/149	Like-for-likeBenchAlign v5.8 lane · 3 vs 10 public rows	Practical tie
Knowledge	59.6Supported · #49/181	49.6Supported · #80/181	Like-for-likeBenchAlign v5.8 lane · 1 vs 2 public rows	GPT-5.2 leads · intervals overlap
Multimodal	75.9#22/54	61.5#43/54	Directional onlyProvisional lane · 2 vs 2 weighted rows	Directional only
Instruction following	92.6#19/144	93.8#5/144	Directional onlyProvisional lane · 0 vs 0 weighted rows	Directional only
Reasoning	64.3Unranked · 4 rankable rows	82.9#5/29	Not comparableProvisional lane · 1 vs 0 weighted rows	Not comparable
Multilingual	Not ranked	Not ranked	Not comparableProvisional lane · 0 vs 0 weighted rows	Not comparable
Math	71.1Unranked · 3 rankable rows	Not ranked	Not comparableProvisional lane · 2 vs 1 weighted rows	Not comparable

Ranks count the models scored in each category’s lane, so the agentic, coding, and knowledge denominators (BenchAlign v5.8) differ from the provisional-lane categories. Unranked scores sit on the provisional lane’s scale but never name a winner. Like-for-like rows within 0.5 points read as a practical tie.

Supported evidence per lane · bars run 0–100Methodology

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-5.2: $0.00875; 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-5.2: $0.1295; 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-5.2: $0.525; Fits in one request; Cached input priced at the published list-input rate
MiniMax M3: $0.03; Fits in one request

MiniMax M3 has the lower modeled cost

GPT-5.2 has no published cached-input rate, so cached tokens use its listed input rate.

Cached input falls back to the list input rate only where a cached rate is unpublished

Specification differences

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

SpecificationGPT-5.2MiniMax M3

Context window

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

GPT-5.2

400K

MiniMax M3

API model ID

GPT-5.2

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.

GPT-5.2

Not published

MiniMax M3

$0.06 per 1M cached input tokens

Documented inputs

GPT-5.2

Not sourced

MiniMax M3

Not sourced

Documented outputs

GPT-5.2

Not sourced

MiniMax M3

Not sourced

Provider availability

GPT-5.2

Not sourced

MiniMax M3

Not sourced

Reasoning profile

GPT-5.2

Reasoning

MiniMax M3

Non-Reasoning

Weight access

GPT-5.2

Proprietary

MiniMax M3

Open Weight

License

GPT-5.2

Proprietary

MiniMax M3

Open Weight

Release date

GPT-5.2

2025-12-11

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-5.2 has the higher public score estimate, 59.85 versus 55, but the 90% score intervals overlap.
Workload cost: Repository review: $0.1295 vs $0.0186. Cache-heavy agent loop: $0.525 vs $0.03.
Context tradeoff: MiniMax M3 has the larger documented window (1M).

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

Questions

Which is better, GPT-5.2 or MiniMax M3?

GPT-5.2 has the higher public score estimate, 59.85 versus 55, 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-5.2 or MiniMax M3?

The like-for-like coding row is a practical tie on the public lane, 36.9 against 36.5, inside the 0.5-point band BenchLM treats as level.

Which is better for agentic tasks, GPT-5.2 or MiniMax M3?

GPT-5.2 has the higher public agentic tasks point estimate, 44.8 to 39.6, with Supported evidence for both models. The conditional ranges do not establish rank confidence.

Which costs less, GPT-5.2 or MiniMax M3?

For the stated presets, chat costs $0.00875 on GPT-5.2 and $0.0009 on MiniMax M3; repository review costs $0.1295 and $0.0186; the cache-heavy agent loop costs $0.525 and $0.03. GPT-5.2 has no published cached-input rate, so cached tokens use its listed input rate.

Which has the larger context window, GPT-5.2 or MiniMax M3?

MiniMax M3 has the larger documented context window: 1M, compared with 400K.

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

BrowseComp
GPT-5.265.8%
Source
MiniMax M383.5%
Source
MiniMax M3 leads this result
OSWorld-Verified
GPT-5.247.3%
Source
MiniMax M370.1%
Source
MiniMax M3 leads this result
Gert Labs
GPT-5.246.54%
Source
MiniMax M3—
Not directly comparable
JobBench
GPT-5.234.3%
Source
MiniMax M3—
Not directly comparable
Terminal-Bench 2.1
GPT-5.2—
MiniMax M366.0%
Source
Not directly comparable
MCP Atlas
GPT-5.2—
MiniMax M374.2%
Source
Not directly comparable
Claw-Eval
GPT-5.2—
MiniMax M374.5%
Source
Not directly comparable
BankerToolBench
GPT-5.2—
MiniMax M376.1%
Source
Not directly comparable
ResearchClawBench
GPT-5.2—
MiniMax M319.8%
Source
Not directly comparable
OSWorld 2.0
GPT-5.2—
MiniMax M34.6%
Source
Not directly comparable
Terminal-Bench 2.1 (Vals)
GPT-5.2—
MiniMax M353.6%
Source
Not directly comparable

Coding

SWE-bench Verified
GPT-5.280%
Source
MiniMax M380.5%
Source
MiniMax M3 leads this result
SWE-bench Pro
GPT-5.255.6%
Source
MiniMax M359%
Source
MiniMax M3 leads this result
Vibe Code Bench
GPT-5.253.50%
Source
MiniMax M3—
Not directly comparable
Terminal-Bench 2.1
GPT-5.2—
MiniMax M366.0%
Source
Not directly comparable
NL2Repo
GPT-5.2—
MiniMax M342.1%
Source
Not directly comparable
VIBE V2
GPT-5.2—
MiniMax M350.1%
Source
Not directly comparable
SVG-Bench
GPT-5.2—
MiniMax M363.7%
Source
Not directly comparable
KernelBench Hard
GPT-5.2—
MiniMax M328.8%
Source
Not directly comparable
OpenHarmony Bench
GPT-5.2—
MiniMax M348.4%
Source
Not directly comparable
LiveCodeBench (Vals)
GPT-5.2—
MiniMax M382.2%
Source
Not directly comparable
SWE-bench (Vals)
GPT-5.2—
MiniMax M375.0%
Source
Not directly comparable

Reasoning

ARC-AGI-2
GPT-5.252.9%
Source
MiniMax M3—
Not directly comparable
ARC-AGI-1
GPT-5.286.17%
Source
MiniMax M3—
Not directly comparable

Multimodal

MMMU
GPT-5.286.7%
Source
MiniMax M3—
Not directly comparable
MMMU-Pro
GPT-5.279.5%
Source
MiniMax M378.1%
Source
GPT-5.2 leads this result
RealWorldQA
GPT-5.283.3%
Source
MiniMax M3—
Not directly comparable
OmniDocBench 1.5
GPT-5.285.7%
Source
MiniMax M391.6%
Source
MiniMax M3 leads this result
Video-MME (with subtitle)
GPT-5.286.0%
Source
MiniMax M385.4%
Source
GPT-5.2 leads this result
Video-MME (w/o subtitle)
GPT-5.285.8%
Source
MiniMax M3—
Not directly comparable
MathVision
GPT-5.283.0%
Source
MiniMax M3—
Not directly comparable
We-Math
GPT-5.279.0%
Source
MiniMax M3—
Not directly comparable
DynaMath
GPT-5.286.8%
Source
MiniMax M3—
Not directly comparable
SimpleVQA
GPT-5.255.8%
Source
MiniMax M3—
Not directly comparable
CC-OCR
GPT-5.270.3%
Source
MiniMax M3—
Not directly comparable
AI2D_TEST
GPT-5.292.2%
Source
MiniMax M3—
Not directly comparable
CountBench
GPT-5.291.9%
Source
MiniMax M3—
Not directly comparable
ERQA
GPT-5.259.8%
Source
MiniMax M3—
Not directly comparable
VideoMMMU
GPT-5.285.9%
Source
MiniMax M384.6%
Source
GPT-5.2 leads this result
MLVU (M-Avg)
GPT-5.285.6%
Source
MiniMax M3—
Not directly comparable
CharXiv
GPT-5.282.1%
Source
MiniMax M3—
Not directly comparable
V*
GPT-5.275.9%
Source
MiniMax M3—
Not directly comparable
OfficeQA Pro
GPT-5.2—
MiniMax M345.1%
Source
Not directly comparable

Knowledge

GPQA
GPT-5.292.4%
Source
MiniMax M3—
Not directly comparable
GPQA Diamond (Vals)
GPT-5.2—
MiniMax M392.7%
Source
Not directly comparable
MMLU-Pro (Vals)
GPT-5.2—
MiniMax M384.2%
Source
Not directly comparable

Math

AIME 2025
GPT-5.2100%
Source
MiniMax M3—
Not directly comparable
FrontierMath v2 (Tiers 1-3)
GPT-5.267.400%
Source
MiniMax M3—
Not directly comparable
FrontierMath v2 (Tier 4)
GPT-5.231.700%
Source
MiniMax M3—
Not directly comparable
USAMO 2026
GPT-5.2—
MiniMax M385.7%
Source
Not directly comparable

50 public results · 8 shared

Related comparisons

Compare API pricing Read the methodology Open the model selector

Watch GPT-5.2 vs MiniMax M3

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

Read a sample issue

Join 5,500+ readers.

Last updated October 14, 2026

GPT-5.2 vs MiniMax M3

Decision reading

Which one for your work

Agentic work

Long documents

Chat turn cost

Cache-heavy agent loop cost

Repository review cost

Coding work

Which one for a specific job

What is actually comparable

Shape of the matched evidence

Category results, on a stated basis

Agentic

Coding

Knowledge

Multimodal

Instruction following

Reasoning

Multilingual

Math

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

If you already use one of these models

Questions

Which is better, GPT-5.2 or MiniMax M3?

Which is better for coding, GPT-5.2 or MiniMax M3?

Which is better for agentic tasks, GPT-5.2 or MiniMax M3?

Which costs less, GPT-5.2 or MiniMax M3?

Which has the larger context window, GPT-5.2 or MiniMax M3?

Benchmark evidence

Agentic

Coding

Reasoning

Multimodal

Knowledge

Math

Related comparisons

Watch GPT-5.2 vs MiniMax M3