Model A

Anthropic

80.66/100

Supported · Public rank #4

90% interval 78.3–83.0

Claude Opus 5 vs MiniMax M3

Updated September 4, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.

Model B

MiniMax M3

MiniMax

63.91/100

Supported · Public rank #51

90% interval 56.5–71.3

Decision reading

Claude Opus 5 has the higher public score, 80.66 versus 63.91, and the 90% score intervals do not overlap.

11 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 X LinkedInSocial card CSV JSON

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
Claude Opus 5
Claude Opus 5 leads on the public agentic lane, 77.4 to 43.2, with Supported evidence for both models and non-overlapping 90% intervals.
Confidence: stronger
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
MiniMax M3 is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like.
Confidence: limited
Long documents
Prompts that approach the documented context limit
No clear pick
The documented context windows are equal.
Confidence: documented

What is actually comparable

Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.

62 Claude Opus 5 only

11 Shared

16 MiniMax M3 only

Shared results: 11
Claude Opus 5 only: 62
MiniMax M3 only: 16
Like-for-like categories: 2 / 8

3 categories rest on Estimated evidence or different evidence sets. Those rows remain visible for coverage context but do not name a winner.

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

Like-for-like

Claude Opus 5: 77.4; Supported · #2/151
MiniMax M3: 43.2; Supported · #106/151
Basis: BenchAlign lane · 19 vs 9 public rows
Reading: Claude Opus 5 leads

Knowledge

Like-for-like

Claude Opus 5: 82.1; Supported · #3/181
MiniMax M3: 53.9; Supported · #65/181
Basis: BenchAlign lane · 19 vs 2 public rows
Reading: Claude Opus 5 leads

Coding

Directional only

Claude Opus 5: 75.6; Supported · #3/183
MiniMax M3: 50.6; Estimated · #68/183
Basis: BenchAlign lane · 16 vs 10 public rows
Reading: Directional only

Reasoning

Directional only

Claude Opus 5: 75.6; #13/22
MiniMax M3: 78.5; #5/22
Basis: Provisional lane · 2 vs 0 weighted rows
Reading: Directional only

Multimodal

Directional only

Claude Opus 5: 88.7; #2/48
MiniMax M3: 52.0; #34/48
Basis: Provisional lane · 1 vs 2 weighted rows
Reading: Directional only

Math

Not comparable

Claude Opus 5: Not ranked
MiniMax M3: Not ranked
Basis: Provisional lane · 0 vs 1 weighted rows
Reading: Not comparable

Multilingual

Not comparable

Claude Opus 5: Not ranked
MiniMax M3: Not ranked
Basis: Provisional lane · 0 vs 0 weighted rows
Reading: Not comparable

Instruction following

Not comparable

Claude Opus 5: Not ranked
MiniMax M3: 93.5; #4/120
Basis: Provisional lane · 0 vs 0 weighted rows
Reading: Not comparable

Category scores with the server-provided evidence basis for Claude Opus 5 and MiniMax M3
Category	Claude Opus 5	MiniMax M3	Basis	Reading
Agentic	77.4Supported · #2/151	43.2Supported · #106/151	Like-for-likeBenchAlign lane · 19 vs 9 public rows	Claude Opus 5 leads
Knowledge	82.1Supported · #3/181	53.9Supported · #65/181	Like-for-likeBenchAlign lane · 19 vs 2 public rows	Claude Opus 5 leads
Coding	75.6Supported · #3/183	50.6Estimated · #68/183	Directional onlyBenchAlign lane · 16 vs 10 public rows	Directional only
Reasoning	75.6#13/22	78.5#5/22	Directional onlyProvisional lane · 2 vs 0 weighted rows	Directional only
Multimodal	88.7#2/48	52.0#34/48	Directional onlyProvisional lane · 1 vs 2 weighted rows	Directional only
Math	Not ranked	Not ranked	Not comparableProvisional lane · 0 vs 1 weighted rows	Not comparable
Multilingual	Not ranked	Not ranked	Not comparableProvisional lane · 0 vs 0 weighted rows	Not comparable
Instruction following	Not ranked	93.5#4/120	Not comparableProvisional lane · 0 vs 0 weighted rows	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.

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 2.0
Agentic
Claude Opus 5: 70.6%MiniMax M3: 4.6%Normalized gap 66.0Claude Opus 5 source MiniMax M3 source
OfficeQA Pro
Multimodal
Claude Opus 5: 66.9%MiniMax M3: 45.1%Normalized gap 21.8Claude Opus 5 source MiniMax M3 source
SWE-bench Pro
Coding
Claude Opus 5: 79.2%MiniMax M3: 59%Normalized gap 20.2Claude Opus 5 source MiniMax M3 source
SWE-bench Verified
Coding
Claude Opus 5: 96%MiniMax M3: 80.5%Normalized gap 15.5Claude Opus 5 source MiniMax M3 source
MMLU-Pro (Vals)
Knowledge
Claude Opus 5: 91.6%MiniMax M3: 84.2%Normalized gap 7.4Claude Opus 5 source MiniMax M3 source

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

Claude Opus 5: $0.0175; 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

Claude Opus 5: $0.325; 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

Claude Opus 5: $0.45; 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.

SpecificationClaude Opus 5MiniMax M3

Context window

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

Claude Opus 5

Anthropic model overview

MiniMax M3

API model ID

Claude Opus 5

claude-opus-5

Anthropic model overview

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.

Claude Opus 5

$0.5 per 1M cached input tokens

Claude API pricing

MiniMax M3

$0.06 per 1M cached input tokens

Documented inputs

Claude Opus 5

text, image

Anthropic model overview

MiniMax M3

Not sourced

Documented outputs

Claude Opus 5

text

Anthropic model overview

MiniMax M3

Not sourced

Provider availability

Claude Opus 5

Generally Available · Claude API

Anthropic model overview

MiniMax M3

Not sourced

Reasoning profile

Claude Opus 5

Reasoning

MiniMax M3

Non-Reasoning

Weight access

Claude Opus 5

Proprietary

MiniMax M3

Open Weight

License

Claude Opus 5

Proprietary

MiniMax M3

Open Weight

Release date

Claude Opus 5

2026-07-24

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: Claude Opus 5 has the higher public score, 80.66 versus 63.91, and the 90% score intervals do not overlap.
Workload cost: Repository review: $0.325 vs $0.0186. Cache-heavy agent loop: $0.45 vs $0.03.
Context tradeoff: Both models list 1M.

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 evidence89 rows

Agentic

Terminal-Bench 3.0
Claude Opus 542.7%
Source
MiniMax M3—
Not directly comparable
BrowseComp
Claude Opus 590.8%
Source
MiniMax M383.5%
Source
Claude Opus 5 leads this result
HLE w/ tools
Claude Opus 564.7%
Source
MiniMax M3—
Not directly comparable
DeepSearchQA
Claude Opus 595.0%
Source
MiniMax M3—
Not directly comparable
DRACO
Claude Opus 588.6%
Source
MiniMax M3—
Not directly comparable
BrowseComp (10-agent, prerelease)
Claude Opus 593.6%
Source
MiniMax M3—
Not directly comparable
OSWorld 2.0
Claude Opus 570.6%
Source
MiniMax M34.6%
Source
Claude Opus 5 leads this result
MCP Atlas
Claude Opus 585.8%
Source
MiniMax M374.2%
Source
Claude Opus 5 leads this result
MCP-Atlas claim coverage
Claude Opus 589.1%
Source
MiniMax M3—
Not directly comparable
LAB all-pass (Anthropic harness)
Claude Opus 523.58%
Source
MiniMax M3—
Not directly comparable
LAB criterion-pass (Anthropic harness)
Claude Opus 593.74%
Source
MiniMax M3—
Not directly comparable
LAB all-pass (Harvey held-out)
Claude Opus 511.7%
Source
MiniMax M3—
Not directly comparable
LAB criterion-pass (Harvey held-out)
Claude Opus 594.1%
Source
MiniMax M3—
Not directly comparable
Toolathlon-Verified
Claude Opus 580.6%
Source
MiniMax M3—
Not directly comparable
Toolathlon Verified Pass@3
Claude Opus 587.0%
Source
MiniMax M3—
Not directly comparable
Toolathlon Verified Pass³
Claude Opus 573.1%
Source
MiniMax M3—
Not directly comparable
Toolathlon Verified avg. turns
Claude Opus 523.5 turns
Source
MiniMax M3—
Not directly comparable
AutomationBench
Claude Opus 526.0%
Source
MiniMax M3—
Not directly comparable
Terminal-Bench 2.1 (Vals)
Claude Opus 584.6%
Source
MiniMax M353.6%
Source
Claude Opus 5 leads this result
Terminal-Bench 2.0
Claude Opus 5—
MiniMax M366%
Source
Not directly comparable
OSWorld-Verified
Claude Opus 5—
MiniMax M370.1%
Source
Not directly comparable
Claw-Eval
Claude Opus 5—
MiniMax M374.5%
Source
Not directly comparable
BankerToolBench
Claude Opus 5—
MiniMax M376.1%
Source
Not directly comparable
ResearchClawBench
Claude Opus 5—
MiniMax M319.8%
Source
Not directly comparable

Coding

Bug Hunt Bench
Claude Opus 527 fixes
Source
MiniMax M3—
Not directly comparable
SWE-bench Verified
Claude Opus 596%
Source
MiniMax M380.5%
Source
Claude Opus 5 leads this result
SWE-bench Pro
Claude Opus 579.2%
Source
MiniMax M359%
Source
Claude Opus 5 leads this result
SWE Multilingual
Claude Opus 589.5%
Source
MiniMax M3—
Not directly comparable
SWE Multimodal
Claude Opus 559.4%
Source
MiniMax M3—
Not directly comparable
deepSwe
Claude Opus 568.8%
Source
MiniMax M3—
Not directly comparable
FrontierCode 1.1 Main
Claude Opus 553.4%
Source
MiniMax M3—
Not directly comparable
FrontierCode 1.1 Extended
Claude Opus 563.6%
Source
MiniMax M3—
Not directly comparable
FrontierSWE v2
Claude Opus 552.0%
Source
MiniMax M3—
Not directly comparable
ProgramBench (episode 1)
Claude Opus 583.0%
Source
MiniMax M3—
Not directly comparable
ProgramBench
Claude Opus 593.0%
Source
MiniMax M3—
Not directly comparable
cursorBench32
Claude Opus 570.0%
Source
MiniMax M3—
Not directly comparable
VulcanBench v3
Claude Opus 587.0%
Source
MiniMax M3—
Not directly comparable
VulcanBench CII v1
Claude Opus 596.4%
Source
MiniMax M3—
Not directly comparable
LiveCodeBench (Vals)
Claude Opus 589.0%
Source
MiniMax M382.2%
Source
Claude Opus 5 leads this result
SWE-bench (Vals)
Claude Opus 597.0%
Source
MiniMax M375.0%
Source
Claude Opus 5 leads this result
Terminal-Bench 2.0
Claude Opus 5—
MiniMax M366.0%
Source
Not directly comparable
NL2Repo
Claude Opus 5—
MiniMax M342.1%
Source
Not directly comparable
VIBE V2
Claude Opus 5—
MiniMax M350.1%
Source
Not directly comparable
SVG-Bench
Claude Opus 5—
MiniMax M363.7%
Source
Not directly comparable
KernelBench Hard
Claude Opus 5—
MiniMax M328.8%
Source
Not directly comparable
OpenHarmony Bench
Claude Opus 5—
MiniMax M348.4%
Source
Not directly comparable

Reasoning

ARC-AGI-1
Claude Opus 597.50%
Source
MiniMax M3—
Not directly comparable
ARC-AGI-2
Claude Opus 590.4%
Source
MiniMax M3—
Not directly comparable
ARC-AGI-3
Claude Opus 530.2%
Source
MiniMax M3—
Not directly comparable

Knowledge

HLE
Claude Opus 564.7%
Source
MiniMax M3—
Not directly comparable
HLE w/o tools
Claude Opus 556.3%
Source
MiniMax M3—
Not directly comparable
HLE-Verified
Claude Opus 554.4%
Source
MiniMax M3—
Not directly comparable
LABBench2
Claude Opus 584.2%
Source
MiniMax M3—
Not directly comparable
HealthBench (raw)
Claude Opus 567.1%
Source
MiniMax M3—
Not directly comparable
HealthBench (length-adjusted)
Claude Opus 557.8%
Source
MiniMax M3—
Not directly comparable
HealthBench Professional
Claude Opus 559.8%
Source
MiniMax M3—
Not directly comparable
HealthBench Professional (raw)
Claude Opus 573.4%
Source
MiniMax M3—
Not directly comparable
BioMysteryBench (human-solvable)
Claude Opus 590.1%
Source
MiniMax M3—
Not directly comparable
BioMysteryBench (human-difficult)
Claude Opus 549.4%
Source
MiniMax M3—
Not directly comparable
SpatialBench Verified
Claude Opus 572.5%
Source
MiniMax M3—
Not directly comparable
SingleCellBench
Claude Opus 560.6%
Source
MiniMax M3—
Not directly comparable
ProteinGym Hard
Claude Opus 547.7%
Source
MiniMax M3—
Not directly comparable
Protein Design
Claude Opus 542.5%
Source
MiniMax M3—
Not directly comparable
Organic chemistry V2
Claude Opus 561.6%
Source
MiniMax M3—
Not directly comparable
Protocols (troubleshooting)
Claude Opus 561.1%
Source
MiniMax M3—
Not directly comparable
Protocols (understanding)
Claude Opus 578.4%
Source
MiniMax M3—
Not directly comparable
GPQA Diamond (Vals)
Claude Opus 593.4%
Source
MiniMax M392.7%
Source
Claude Opus 5 leads this result
MMLU-Pro (Vals)
Claude Opus 591.6%
Source
MiniMax M384.2%
Source
Claude Opus 5 leads this result

Math

IMO 2026
Claude Opus 542/42
Source
MiniMax M3—
Not directly comparable
RiemannBench (no tools)
Claude Opus 560.0%
Source
MiniMax M3—
Not directly comparable
RiemannBench (tools)
Claude Opus 579.0%
Source
MiniMax M3—
Not directly comparable
ArXivMath Jun. 2026 (no tools)
Claude Opus 590.8%
Source
MiniMax M3—
Not directly comparable
ArXivMath Jun. 2026 (tools)
Claude Opus 591.3%
Source
MiniMax M3—
Not directly comparable
USAMO 2026
Claude Opus 5—
MiniMax M385.7%
Source
Not directly comparable

Multilingual

GMMLU
Claude Opus 592.5%
Source
MiniMax M3—
Not directly comparable
MILU
Claude Opus 592.1%
Source
MiniMax M3—
Not directly comparable
INCLUDE
Claude Opus 589.8%
Source
MiniMax M3—
Not directly comparable

Multimodal

Chartography (no tools)
Claude Opus 529.6%
Source
MiniMax M3—
Not directly comparable
Chartography (tools)
Claude Opus 583.0%
Source
MiniMax M3—
Not directly comparable
BenchCAD Vision2Code (no tools)
Claude Opus 50.366
Source
MiniMax M3—
Not directly comparable
BenchCAD Vision2Code (tools)
Claude Opus 50.821
Source
MiniMax M3—
Not directly comparable
GDP.pdf (no tools)
Claude Opus 583.4%
Source
MiniMax M3—
Not directly comparable
GDP.pdf (tools)
Claude Opus 585.5%
Source
MiniMax M3—
Not directly comparable
OfficeQA
Claude Opus 578.1%
Source
MiniMax M3—
Not directly comparable
OfficeQA Pro
Claude Opus 566.9%
Source
MiniMax M345.1%
Source
Claude Opus 5 leads this result
OmniDocBench 1.5
Claude Opus 5—
MiniMax M391.6%
Source
Not directly comparable
MMMU-Pro
Claude Opus 5—
MiniMax M378.1%
Source
Not directly comparable
VideoMMMU
Claude Opus 5—
MiniMax M384.6%
Source
Not directly comparable
Video-MME (with subtitle)
Claude Opus 5—
MiniMax M385.4%
Source
Not directly comparable

Frequently asked questions

Which is better, Claude Opus 5 or MiniMax M3?

Claude Opus 5 has the higher public score, 80.66 versus 63.91, 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, Claude Opus 5 or MiniMax M3?

Claude Opus 5 scores higher for coding on the public lane, 75.6 to 50.6. MiniMax M3 is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like. BenchLM does not name a winner for coding from a directional reading; read the shared benchmark rows directly and test the models on the same task set.

Which is better for agentic tasks, Claude Opus 5 or MiniMax M3?

Claude Opus 5 leads the public agentic tasks lane, 77.4 to 43.2, with Supported evidence for both models and non-overlapping 90% intervals.

Which costs less, Claude Opus 5 or MiniMax M3?

For the stated presets, chat costs $0.0175 on Claude Opus 5 and $0.0009 on MiniMax M3; repository review costs $0.325 and $0.0186; the cache-heavy agent loop costs $0.45 and $0.03. Costs use the listed standard API rates.

Which has the larger context window, Claude Opus 5 or MiniMax M3?

Both models list the same context window, 1M.

Related comparisons

Compare API pricing Read the methodology Open the model selector

Last updated September 4, 2026

Watch Claude Opus 5 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.

Claude Opus 5 vs MiniMax M3

Which one for your work

Agentic work

Chat turn cost

Cache-heavy agent loop cost

Repository review cost

Coding work

Long documents

What is actually comparable

Category results, on a stated basis

Agentic

Knowledge

Coding

Reasoning

Multimodal

Math

Multilingual

Instruction following

Shape of the matched evidence

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

Benchmark evidence

Agentic

Coding

Reasoning

Knowledge

Math

Multilingual

Multimodal

Frequently asked questions

Which is better, Claude Opus 5 or MiniMax M3?

Which is better for coding, Claude Opus 5 or MiniMax M3?

Which is better for agentic tasks, Claude Opus 5 or MiniMax M3?

Which costs less, Claude Opus 5 or MiniMax M3?

Which has the larger context window, Claude Opus 5 or MiniMax M3?

Related comparisons

Watch Claude Opus 5 vs MiniMax M3