MiniMax-M3 Intelligence, Performance & Price Analysis
Model summary
IntelligenceUpdated
Speed
Price
Cache Price
Verbosity
MiniMax-M3 is amongst the leading models in intelligence and reasonably priced when comparing to other open weight models of similar size. It's also faster than average and highly concise. The model supports text, image, and video input, outputs text, and has a 1m tokens context window.
MiniMax-M3 scores 44 on the Artificial Analysis Intelligence Index, placing it well above average among comparable models (averaging 24). When evaluating the Intelligence Index, it generated 89M tokens, which is very concise in comparison to the average of 110M.
Pricing for MiniMax-M3 is $0.30 per 1M input tokens (moderately priced, average: $0.42) and $1.20 per 1M output tokens (moderately priced, average: $1.25). In total, it cost $203.86 to evaluate MiniMax-M3 on the Intelligence Index.
At 85 tokens per second, MiniMax-M3 is faster than average (72).
| Reasoning | Yes This page shows the reasoning version of this model. A non-reasoning variant may also exist. |
|---|---|
| Input modality | Supports: text, image, video |
| Output modality | Supports: text |
| Context window | 1m ~1500 A4 pages of size 12 Arial font |
| Total parameters | 428B |
| Active parameters | 23B Number of parameters active per token during inference |
| License | MINIMAX COMMUNITY LICENSE |
| Model weights | Hugging Face |
Metrics are compared against models of the same class:
- Non-reasoning models → compared only with other non-reasoning models
- Reasoning models → compared across both reasoning and non-reasoning
- Open weights models → compared only with other open weights models of the same size class:
- Tiny: ≤4B parameters
- Small: 4B–40B parameters
- Medium: 40B–150B parameters
- Large: >150B parameters
- Proprietary models → compared across proprietary and open weights models of the same price range, using a blended 3:1 input/output price ratio:
- <$0.15 per 1M tokens
- $0.15–$1 per 1M tokens
- >$1 per 1M tokens
Highlights
IntelligenceUpdated
Artificial Analysis Intelligence Index
Artificial Analysis Intelligence Index by Open Weights / Proprietary
Intelligence Evaluations
Agentic real-world work tasks, (Elo-500)/2000
Agentic tool use
Agentic coding & terminal use
Coding
Reasoning & knowledge
Scientific reasoning
Physics reasoning
Knowledge
1 - hallucination rate
Long context reasoning
Instruction following
Long-horizon agentic tasks
Kubernetes incident root-cause analysis
Visual reasoning
AA-BriefcaseNew
AA-Briefcase Elo
Openness
Artificial Analysis Openness Index: Score
Intelligence Index Comparisons
Intelligence vs. Cost per Intelligence Index Task
Token UseUpdated
Output Tokens per Intelligence Index Task
Price and CostUpdated
Cost per Intelligence Index Task
Cost to Run Artificial Analysis Intelligence Index
Pricing: Cache Hit, Input, and Output
Context Window
Context Window
SpeedUpdated
Measured by Output Speed (tokens per second)
Output Speed
Time per Intelligence Index Task
Latency
Measured by Time (seconds) to First Token
Latency: Time To First Answer Token
End-to-End Response Time
Seconds to output 500 tokens, calculated based on time to first token, 'thinking' time for reasoning models, and output speed
End-to-End Response Time
Model Size (Open Weights Models Only)
Model Size: Total and Active Parameters
Frequently Asked Questions
Common questions about MiniMax-M3
MiniMax-M3 was released on June 1, 2026.
MiniMax-M3 was created by MiniMax.
MiniMax-M3 scores 44 on the Artificial Analysis Intelligence Index, placing it well above average among other open weight models of similar size (median: 24).
MiniMax-M3 generates output at 85.4 tokens per second (based on MiniMax's API), which is above average compared to other open weight models of similar size (median: 72.1 t/s).
MiniMax-M3 has a time to first token (TTFT) of 3.49s (based on MiniMax's API), which is at the higher end compared to other open weight models of similar size (median: 2.40s).
MiniMax-M3 costs $0.30 per 1M input tokens (very competitive, median: $0.55) and $1.20 per 1M output tokens (better than average, median: $1.85), based on MiniMax's API.
MiniMax-M3 costs $0.30 per 1M input tokens and $1.20 per 1M output tokens (based on MiniMax's API). For a blended rate (7:2:1 cache hit/input/output ratio), this is $0.22 per 1M tokens. Pricing may vary by provider. Compare provider pricing
When evaluated on the Intelligence Index, MiniMax-M3 generated 89M output tokens, which is very competitive compared to other open weight models of similar size (median: 110M).
Yes, MiniMax-M3 is a reasoning model. It uses extended thinking or chain-of-thought reasoning to work through complex problems before providing an answer.
MiniMax-M3 supports text, image, and video input.
MiniMax-M3 supports text output.
Yes, MiniMax-M3 supports image input and can analyze, describe, and answer questions about images.
Yes, MiniMax-M3 is multimodal. It can process text, image, and video input and generate text output.
MiniMax-M3 has a context window of 1.0M tokens. This determines how much text and conversation history the model can process in a single request.
Yes, MiniMax-M3 is open weights. The model weights are publicly available and can be downloaded for self-hosting.
MiniMax-M3 has 428 billion parameters (23 billion active).
MiniMax-M3 is a Mixture of Experts (MoE) model with 428 billion total parameters, but only 23 billion active parameters are used during inference.
MiniMax-M3 is released under the MINIMAX COMMUNITY LICENSE license. Commercial use requires a separate license agreement. View license
MiniMax-M3 achieves a score of 44 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.
MiniMax-M3 is an open weights model that can be self-hosted. View providers
MiniMax-M3 is an open weights model that can be downloaded and self-hosted. Compare providers