MiniMax model product
MiniMax M3
MiniMax M3 is an open-weight LLM with native image and video input, desktop computer operation, and a 1M-token context window.
Updated Oct 23, 2026. Default version: MiniMax M3
LLMBoard Score68.3MiniMax M3
Coverage80%25 benchmark families
Context window1MTokens
Official input price$0.30MiniMax (minimax.io) API
MiniMax M3 Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
MiniMax M3 Benchmark Results
Benchmark scores for MiniMax M3.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkGDPval-Rubrics | Score74.78% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkIMO 2025 | Score35.00 points | Rank01 | Participants2 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkKernelBench Hard | Score28.80% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkLiveSQLBench | Score40.17% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkOmniDocBench 1.5 | Score91.60% | Rank01 | Participants20 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkSpreadSheetBench-v1 | Score89.35% | Rank01 | Participants4 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkSVG-Bench | Score63.70% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkSWE Atlas - Test Writing | Score30.83% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkSWE-fficiency | Score34.80% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkUSAMO 2026 | Score36.00 points | Rank01 | Participants3 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkVIBE-V2 | Score50.12% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkYC-Bench | Score2,100,000.00 points | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkCL-bench | Score20.48% | Rank02 | Participants3 | Percentile50.00% | EvidenceC | Evaluated |
| BenchmarkLOCA-Bench (256k) | Score49.30% | Rank02 | Participants2 | Percentile0.00% | EvidenceC | Evaluated |
| BenchmarkAA IFBench | Score82.86% | Rank02 | Participants168 | Percentile99.40% | EvidenceB | EvaluatedN/A |
| BenchmarkBankerToolBench | Score76.12% | Rank03 | Participants3 | Percentile0.00% | EvidenceC | Evaluated |
| BenchmarkClaw-Eval | Score74.50% | Rank03 | Participants14 | Percentile84.62% | EvidenceC | Evaluated |
| BenchmarkDRACO | Score73.23% | Rank03 | Participants4 | Percentile33.33% | EvidenceC | Evaluated |
| BenchmarkPaperBench | Score52.60% | Rank03 | Participants3 | Percentile0.00% | EvidenceC | Evaluated |
| BenchmarkPostTrainBench | Score37.10% | Rank04 | Participants9 | Percentile62.50% | EvidenceC | Evaluated |
| BenchmarkSWE Atlas - Codebase QnA | Score37.90% | Rank06 | Participants7 | Percentile16.67% | EvidenceC | Evaluated |
| BenchmarkVideo-MME | Score85.40% | Rank06 | Participants17 | Percentile68.75% | EvidenceC | Evaluated |
| BenchmarkVideoMMMU | Score84.60% | Rank08 | Participants28 | Percentile74.07% | EvidenceC | Evaluated |
| BenchmarkAPEX-Agents | Score27.70% | Rank10 | Participants11 | Percentile10.00% | EvidenceC | Evaluated |
| BenchmarkAA LCR v1.1 | Score83.00% | Rank11 | Participants193 | Percentile94.79% | EvidenceB | EvaluatedN/A |
| BenchmarkFinance Agent v2 | Score48.27% | Rank12 | Participants29 | Percentile60.71% | EvidenceB | Evaluated |
| BenchmarkOfficeQA Pro | Score45.10% | Rank12 | Participants12 | Percentile0.00% | EvidenceC | Evaluated |
| BenchmarkSWE-Bench Verified | Score80.50% | Rank12 | Participants120 | Percentile90.76% | EvidenceC | Evaluated |
| BenchmarkAA GPQA Diamond | Score92.93% | Rank13 | Participants199 | Percentile93.94% | EvidenceB | EvaluatedN/A |
| BenchmarkBrowseComp | Score83.52% | Rank18 | Participants70 | Percentile75.36% | EvidenceC | Evaluated |
MiniMax M3 Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenaagent bash reliability no prereq | Categoryoverall | Rank24 | Rating / score0.02 | VotesN/A | Observations33.1K | Result date |
| Arenaagent bash recovery steps | Categoryoverall | Rank25 | Rating / score0.01 | VotesN/A | Observations89.5K | Result date |
| Arenadocument | Categoryoverall | Rank27 | Rating / score1,434.79 | Votes6,314 | ObservationsN/A | Result date |
| Arenatext factuality | Categorygerman | Rank28 | Rating / score1,406.46 | Votes566 | ObservationsN/A | Result date |
| Arenatext factuality | Categorypolish | Rank29 | Rating / score1,428.65 | Votes610 | ObservationsN/A | Result date |
| Arenaagent bash reliability | Categoryoverall | Rank29 | Rating / score0.00 | VotesN/A | Observations50.8K | Result date |
| Arenavision | Categoryentity recognition | Rank31 | Rating / score1,196.75 | Votes107 | ObservationsN/A | Result date |
| Arenadocument style control | Categoryoverall | Rank33 | Rating / score1,432.96 | Votes6,314 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryjapanese | Rank33 | Rating / score1,390.84 | Votes619 | ObservationsN/A | Result date |
| Arenawebdev | Categoryimage to webdev | Rank34 | Rating / score1,470.34 | Votes5,983 | ObservationsN/A | Result date |
| Arenaagent steerability | Categoryoverall | Rank35 | Rating / score-0.10 | VotesN/A | Observations64.4K | Result date |
| Arenavision | Categorychinese | Rank36 | Rating / score1,311.08 | Votes1,048 | ObservationsN/A | Result date |
| Arenavision style control | Categoryentity recognition | Rank36 | Rating / score1,160.85 | Votes107 | ObservationsN/A | Result date |
| Arenaagent tool hallucination | Categoryoverall | Rank38 | Rating / score-0.02 | VotesN/A | Observations3.3M | Result date |
| Arenatext factuality | Categorykorean | Rank38 | Rating / score1,380.02 | Votes668 | ObservationsN/A | Result date |
| Arenaagent | Categoryoverall | Rank39 | Rating / score-0.10 | VotesN/A | Observations3.4M | Result date |
| Arenaagent task outcome explicit | Categoryoverall | Rank39 | Rating / score-0.16 | VotesN/A | Observations27.2K | Result date |
| Arenaagent praise complaint | Categoryoverall | Rank40 | Rating / score-0.16 | VotesN/A | Observations12K | Result date |
| Arenavision style control | Categorychinese | Rank40 | Rating / score1,289.32 | Votes1,048 | ObservationsN/A | Result date |
| Arenavision | Categoryenglish | Rank44 | Rating / score1,262.56 | Votes7,372 | ObservationsN/A | Result date |
| Arenavision | Categorydiagram | Rank47 | Rating / score1,266.36 | Votes5,289 | ObservationsN/A | Result date |
| Arenavision | Categoryhumor | Rank47 | Rating / score1,235.03 | Votes773 | ObservationsN/A | Result date |
| Arenavision | Categoryocr | Rank47 | Rating / score1,258.56 | Votes13,947 | ObservationsN/A | Result date |
| Arenawebdev | Categoryoverall | Rank47 | Rating / score1,482.37 | Votes19,803 | ObservationsN/A | Result date |
| Arenawebdev | Categorywebdev | Rank47 | Rating / score1,482.37 | Votes19,803 | ObservationsN/A | Result date |
| Arenawebdev | Categorywebdev-react | Rank47 | Rating / score1,478.08 | Votes13,888 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry mathematical | Rank48 | Rating / score1,447.21 | Votes3,181 | ObservationsN/A | Result date |
| Arenavision | Categoryoverall | Rank48 | Rating / score1,251.05 | Votes19,844 | ObservationsN/A | Result date |
| Arenavision style control | Categoryenglish | Rank49 | Rating / score1,242.71 | Votes7,372 | ObservationsN/A | Result date |
| Arenavision style control | Categoryhumor | Rank49 | Rating / score1,213.43 | Votes773 | ObservationsN/A | Result date |
MiniMax M3 Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $0.30 input, $1.2 output per 1M
- Official provider
- MiniMax (minimax.io)
- Lowest third-party
- From $0.20 input, $0.90 output per 1M via Vultr
- Tracked offerings
- 56
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderMiniMax Token Plan (minimax.cn) | Provider model IDMiniMax-M3 | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context1M | Updated |
| ProviderVolcengine Ark Coding Plan | Provider model IDminimax-m3 | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context1M | Updated |
| ProviderMiniMax Token Plan (minimax.io) | Provider model IDMiniMax-M3 | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context1M | Updated |
| ProviderSCNet Token Plan | Provider model IDMiniMax-M3 | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context1M | Updated |
| ProviderVultr | Provider model IDminimax-m3 | Regionglobal | Input / 1M$0.20 | Output / 1M$0.90 | Context524.3K | Updated |
| ProviderKilo Gateway | Provider model IDminimax/minimax-m3 | Regionglobal | Input / 1M$0.224 | Output / 1M$0.88 | Context524.3K | Updated |
| ProviderEmpirioLabs AI | Provider model IDminimax-m3 | Regionglobal | Input / 1M$0.225 | Output / 1M$0.90 | Context1M | Updated |
| ProviderNanoGPT | Provider model IDminimax/minimax-m3 | Regionglobal | Input / 1M$0.23 | Output / 1M$0.96 | Context512K | Updated |
| ProviderCoreWeave | Provider model IDMiniMaxAI/MiniMax-M3 | Regionglobal | Input / 1M$0.23 | Output / 1M$0.96 | Context262.1K | Updated |
| ProviderAmbient | Provider model IDminimax/minimax-m3 | Regionglobal | Input / 1M$0.23 | Output / 1M$0.96 | Context262.1K | Updated |
| ProviderVancine | Provider model IDMiniMax-M3 | Regionglobal | Input / 1M$0.24 | Output / 1M$0.96 | Context1M | Updated |
| ProviderDeep Infra | Provider model IDMiniMaxAI/MiniMax-M3 | Regionglobal | Input / 1M$0.28 | Output / 1M$1.1 | Context524.3K | Updated |
| ProviderLilac | Provider model IDminimaxai/minimax-m3 | Regionglobal | Input / 1M$0.28 | Output / 1M$1.1 | Context1M | Updated |
| ProviderDevPass (LLM Gateway) | Provider model IDminimax-m3 | Regionglobal | Input / 1M$0.28 | Output / 1M$1.1 | Context1M | Updated |
| ProviderBoundless | Provider model IDminimax-m3 | Regionglobal | Input / 1M$0.28 | Output / 1M$1.1 | Context1M | Updated |
| ProviderAIHubMix | Provider model IDminimax-m3 | Regionglobal | Input / 1M$0.288 | Output / 1M$1.15 | Context1M | Updated |
| ProviderOrcaRouter | Provider model IDminimax/minimax-m3 | Regionglobal | Input / 1M$0.30 | Output / 1M$1.2 | Context1M | Updated |
| ProviderCortecs | Provider model IDminimax-m3 | Regionglobal | Input / 1M$0.30 | Output / 1M$1.2 | Context1M | Updated |
| ProviderJalapeno Cloud | Provider model IDMiniMax-M3 | Regionglobal | Input / 1M$0.30 | Output / 1M$1.2 | Context524.3K | Updated |
| ProviderHugging Face | Provider model IDMiniMaxAI/MiniMax-M3 | Regionglobal | Input / 1M$0.30 | Output / 1M$1.2 | Context524.3K | Updated |
| ProviderOpenRouter | Provider model IDminimax/minimax-m3 | Regionglobal | Input / 1M$0.30 | Output / 1M$1.2 | Context1M | Updated |
| ProviderOpenCode Go | Provider model IDminimax-m3 | Regionglobal | Input / 1M$0.30 | Output / 1M$1.2 | Context1M | Updated |
| ProviderNebius Token Factory | Provider model IDMiniMaxAI/MiniMax-M3 | Regionglobal | Input / 1M$0.30 | Output / 1M$1.2 | Context1M | Updated |
| ProviderMerge Gateway | Provider model IDminimax/minimax-m3 | Regionglobal | Input / 1M$0.30 | Output / 1M$1.2 | Context1M | Updated |
| ProviderZenMux | Provider model IDminimax/minimax-m3 | Regionglobal | Input / 1M$0.30 | Output / 1M$1.2 | Context1M | Updated |
| ProviderTogether AI | Provider model IDMiniMaxAI/MiniMax-M3 | Regionglobal | Input / 1M$0.30 | Output / 1M$1.2 | Context524.3K | Updated |
| ProviderOpenCode | Provider model IDminimax-m3 | Regionglobal | Input / 1M$0.30 | Output / 1M$1.2 | Context512K | Updated |
| ProviderRequesty | Provider model IDminimax-m3 | Regionglobal | Input / 1M$0.30 | Output / 1M$1.2 | Context1M | Updated |
| ProviderClinePass | Provider model IDcline-pass/minimax-m3 | Regionglobal | Input / 1M$0.30 | Output / 1M$1.2 | Context1M | Updated |
| ProviderVercel AI Gateway | Provider model IDminimax/minimax-m3 | Regionglobal | Input / 1M$0.30 | Output / 1M$1.2 | Context512K | Updated |
MiniMax M3 Runtime Performance
Provider-specific output speed and catalog latency for MiniMax M3. Runtime does not affect the capability score.
Sort by
Provider | Output Speed | Catalog Latency | Max Input | Max Output | Updated |
|---|
| ProviderMiniMax | Output Speed45.04 tok/s | Catalog Latency11.24 s | Max Input1M | Max Output524.3K | Updated |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
MiniMax M3 Specifications
Technical details for the model's default version.
- Version
- MiniMax M3
- Released
- Jun 1, 2026
- Knowledge cutoff
- Unknown
- Parameters
- 428B
- Context window
- 1M
- Max output
- 524.3K
- Inputs
- image, text, video
- Outputs
- text
- Open weights
- Yes
- License
- MIT
MiniMax M3 Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionMiniMax M3 | Released | LLMBoard68.32 | Parameters428B | Context1M | Max output524.3K | Open weightsYes | LicenseMIT |
Models similar to MiniMax M3
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
MI
MiniMax M3.1 Flash
MiniMax
MI
MiniMax M2.7
MiniMax
MI
MiniMax M2.5
MiniMax
MI
MiniMax M2.1
MiniMax
MI
MiniMax M2
MiniMax
MI
MiniMax M1 80K
MiniMax
What is MiniMax M3?
Key information about MiniMax M3 and its available data.
MiniMax M3 is an open-weight large language model from MiniMax powered by MiniMax Sparse Attention (MSA). It supports native image and video input, desktop computer operation, and toggling thinking on or off at request time.
Data as of 2026-10-22.
FAQ
Common questions about MiniMax M3.
When was MiniMax M3 released?
MiniMax M3's default version was released on Jun 1, 2026.
How much does MiniMax M3 cost?
MiniMax M3's official API price is $0.30 per million input tokens and $1.2 per million output tokens via MiniMax (minimax.io). The lowest tracked third-party offer starts at $0.20 input and $0.90 output via Vultr.
Who created MiniMax M3?
MiniMax M3 was created by MiniMax.
What is the context window for MiniMax M3?
The default version has a 1M token context window.
Is MiniMax M3 open weight?
Yes. The default version is marked as open weight under MIT.
How many API providers offer MiniMax M3?
56 provider offerings are linked to the default version.