Agentic
Not comparable- MiniMax M3
- 42.0
- Supported · #112/154
- Pareto 26.9
- Not ranked
- Basis
- BenchAlign lane · 9 vs 1 public rows
- Reading
- Not comparable
MiniMax M3 vs Pareto 26.9
Decision reading
At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner.
1 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.
Updated September 18, 2026. We do not rank this pair: at least one has no public score. Public scores include evidence status and uncertainty.
Both of these models will change. Get the price, version and retirement notices for the pair, each with its source. Follow model changes
Share or export
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.
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
Pareto 26.9 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
Pareto 26.9 is not ranked on the public lane for agentic, so no winner is named for agentic.
Confidence: limited
Long documents
Prompts that approach the documented context limit
Not enough matched evidence
A complete context comparison is not sourced.
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.
26 MiniMax M3 only
1 Shared
3 Pareto 26.9 only
- Shared results
- 1
- MiniMax M3 only
- 26
- Pareto 26.9 only
- 3
- 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.
Coding
Not comparable- MiniMax M3
- 48.6
- Estimated · #69/154
- Pareto 26.9
- Not ranked
- Basis
- BenchAlign lane · 10 vs 1 public rows
- Reading
- Not comparable
Reasoning
Not comparable- MiniMax M3
- 78.0
- #4/20
- Pareto 26.9
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Knowledge
Not comparable- MiniMax M3
- 53.0
- Supported · #65/184
- Pareto 26.9
- Not ranked
- Basis
- BenchAlign lane · 2 vs 1 public rows
- Reading
- Not comparable
Math
Not comparable- MiniMax M3
- Not ranked
- Pareto 26.9
- Not ranked
- Basis
- Provisional lane · 1 vs 0 weighted rows
- Reading
- Not comparable
Multilingual
Not comparable- MiniMax M3
- Not ranked
- Pareto 26.9
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Multimodal
Not comparable- MiniMax M3
- 52.1
- #34/48
- Pareto 26.9
- Not ranked
- Basis
- Provisional lane · 2 vs 1 weighted rows
- Reading
- Not comparable
Instruction following
Not comparable- MiniMax M3
- 92.4
- #4/124
- Pareto 26.9
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
| Category | MiniMax M3 | Pareto 26.9 | Basis | Reading |
|---|---|---|---|---|
| Agentic | 42.0Supported · #112/154 | Not ranked | Not comparableBenchAlign lane · 9 vs 1 public rows | Not comparable |
| Coding | 48.6Estimated · #69/154 | Not ranked | Not comparableBenchAlign lane · 10 vs 1 public rows | Not comparable |
| Reasoning | 78.0#4/20 | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Knowledge | 53.0Supported · #65/184 | Not ranked | Not comparableBenchAlign lane · 2 vs 1 public rows | Not comparable |
| Math | Not ranked | Not ranked | 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 |
| Multimodal | 52.1#34/48 | Not ranked | Not comparableProvisional lane · 2 vs 1 weighted rows | Not comparable |
| Instruction following | 92.4#4/124 | Not ranked | 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.
MMMU-Pro
Multimodal
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
- MiniMax M3
- $0.0009
- Fits in one request
- Pareto 26.9
- $0.00625
- Fit state unavailable
MiniMax M3 has the lower modeled cost
Costs use the listed standard API rates.
Repository review
50K fresh input + 3K output tokens
- MiniMax M3
- $0.0186
- Fits in one request
- Pareto 26.9
- $0.1475
- Fit state unavailable
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
- MiniMax M3
- $0.03
- Fits in one request
- Pareto 26.9
- $0.175
- Fit state unavailable
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.
SpecificationMiniMax M3Pareto 26.9
Context window
Maximum documented context; output-token limits may be lower.
MiniMax M3
1M
Pareto 26.9
N/A
API model ID
MiniMax M3
Not sourced
Pareto 26.9
pareto
Unbiased pricingCached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
MiniMax M3
$0.06 per 1M cached input tokens
Pareto 26.9
$0.25 per 1M cached input tokens
Unbiased pricingDocumented inputs
MiniMax M3
Not sourced
Pareto 26.9
Not sourced
Documented outputs
MiniMax M3
Not sourced
Pareto 26.9
Not sourced
Provider availability
MiniMax M3
Not sourced
Pareto 26.9
Not sourced
Reasoning profile
MiniMax M3
Non-Reasoning
Pareto 26.9
Reasoning
Weight access
MiniMax M3
Open Weight
Pareto 26.9
Proprietary
License
MiniMax M3
Open Weight
Pareto 26.9
Proprietary
Release date
MiniMax M3
2026-06-01
Pareto 26.9
2026-09-17
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
- At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner.
- Workload cost
- Repository review: $0.0186 vs $0.1475. Cache-heavy agent loop: $0.03 vs $0.175.
- Context tradeoff
- A complete documented context comparison is not available.
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 evidence30 rows
Agentic
Terminal-Bench 2.0
MiniMax M366%SourcePareto 26.9—Not directly comparable
BrowseComp
MiniMax M383.5%SourcePareto 26.9—Not directly comparable
OSWorld-Verified
MiniMax M370.1%SourcePareto 26.9—Not directly comparable
MCP Atlas
MiniMax M374.2%SourcePareto 26.9—Not directly comparable
Claw-Eval
MiniMax M374.5%SourcePareto 26.9—Not directly comparable
BankerToolBench
MiniMax M376.1%SourcePareto 26.9—Not directly comparable
ResearchClawBench
MiniMax M319.8%SourcePareto 26.9—Not directly comparable
OSWorld 2.0
MiniMax M34.6%SourcePareto 26.9—Not directly comparable
Terminal-Bench 2.1 (Vals)
MiniMax M353.6%SourcePareto 26.9—Not directly comparable
Terminal-Bench 4.0
MiniMax M3—Pareto 26.951.00%SourceNot directly comparable
Coding
SWE-bench Verified
MiniMax M380.5%SourcePareto 26.9—Not directly comparable
SWE-bench Pro
MiniMax M359%SourcePareto 26.9—Not directly comparable
Terminal-Bench 2.0
MiniMax M366.0%SourcePareto 26.9—Not directly comparable
NL2Repo
MiniMax M342.1%SourcePareto 26.9—Not directly comparable
VIBE V2
MiniMax M350.1%SourcePareto 26.9—Not directly comparable
SVG-Bench
MiniMax M363.7%SourcePareto 26.9—Not directly comparable
KernelBench Hard
MiniMax M328.8%SourcePareto 26.9—Not directly comparable
OpenHarmony Bench
MiniMax M348.4%SourcePareto 26.9—Not directly comparable
LiveCodeBench (Vals)
MiniMax M382.2%SourcePareto 26.9—Not directly comparable
SWE-bench (Vals)
MiniMax M375.0%SourcePareto 26.9—Not directly comparable
DeepSWE
MiniMax M3—Pareto 26.974.0%SourceNot directly comparable
Knowledge
Math
USAMO 2026
MiniMax M385.7%SourcePareto 26.9—Not directly comparable
Multimodal
OfficeQA Pro
MiniMax M345.1%SourcePareto 26.9—Not directly comparable
OmniDocBench 1.5
MiniMax M391.6%SourcePareto 26.9—Not directly comparable
MMMU-Pro
MiniMax M378.1%SourcePareto 26.978%SourceMiniMax M3 leads this result
VideoMMMU
MiniMax M384.6%SourcePareto 26.9—Not directly comparable
Video-MME (with subtitle)
MiniMax M385.4%SourcePareto 26.9—Not directly comparable
Questions
Which is better, MiniMax M3 or Pareto 26.9?
At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner. The page therefore keeps the decision tied to the specific documented workload.
Which is better for coding, MiniMax M3 or Pareto 26.9?
Pareto 26.9 is not ranked on the public lane for coding, so no winner is named for coding.
Which is better for agentic tasks, MiniMax M3 or Pareto 26.9?
Pareto 26.9 is not ranked on the public lane for agentic tasks, so no winner is named for agentic tasks.
Which costs less, MiniMax M3 or Pareto 26.9?
For the stated presets, chat costs $0.0009 on MiniMax M3 and $0.00625 on Pareto 26.9; repository review costs $0.0186 and $0.1475; the cache-heavy agent loop costs $0.03 and $0.175. Costs use the listed standard API rates.
Which has the larger context window, MiniMax M3 or Pareto 26.9?
A complete documented context-window comparison is not available.
Related comparisons
Last updated September 18, 2026
Watch MiniMax M3 vs Pareto 26.9
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.